SECTION III SPECIAL POPULATIONS

Transcription

1 SECTION III SPECIAL POPULATIONS

2 CHAPTER Assessment of Hearing Loss in Children Allan O. Diefendorf INTRODUCTION The current trend in public health and primary health care practice is to view infants and young children suspected of or at risk for hearing loss (see Appendix 23.1) as a high priority for diagnostic evaluation and confirmation of hearing status. This emphasis in hearing health care represents a standard of service delivery that has evolved over the past 35 years of advocacy by the Joint Committee on Infant Hearing (JCIH 1972; 1982; 1991; 1994; 2000; 2007). Specifically, the JCIH has advocated that universal detection has as its goal that 100% of infants with significant congenital hearing loss shall be identified by 3 months of age and shall have appropriate intervention initiated by 6 months of age. To achieve beneficial outcomes for children who are hard of hearing and deaf, an audiologist must provide comprehensive diagnostic evaluation of hearing status within weeks of referral. The diagnostic evaluation provides the first opportunity for developing a relationship with the family and for initiating audiologic care (diagnosis, counseling, intervention, and ongoing care coordination). The interaction with the family during the diagnostic evaluation is critical because the support, guidance, and education a family receives at this time help to facilitate smooth transitions between referral source and early intervention programs in a timely manner for the family and lead to higher rates of compliance with audiologic recommendations. During the diagnostic process, the integrity of the auditory system is evaluated for each ear, and the status of hearing sensitivity across the speech frequency range is described, as is the type (nature) of hearing loss. In turn, these data provide essential information for medical management when indicated, as well as the data required to initiate amplification protocols. Finally, these data are further used as a baseline for continued audiologic monitoring. The purpose of this chapter is to focus on the audiologic assessment and diagnosis of infants and young children with hearing loss. Additionally, this chapter is developed with an emphasis on the following premise: audiologic procedures must be age appropriate, outcome based, and cost effective, and all procedures must have demonstrated validity and reliability. JUSTIFICATION FOR EARLY DETECTION OF HEARING LOSS Undetected hearing loss in infants and young children compromises optimal language development and personal achievement. Without appropriate opportunities to learn language, children will fall behind their hearing peers in language, cognition, social-emotional development, and academic achievement. However, research demonstrates that when hearing loss is identified early (prior to 6 months of age) and followed immediately (within 2 months) with appropriate intervention services, the outcomes in language development, speech development, and social-emotional development will be significantly better when compared with children with later identified congenital hearing loss (Yoshinaga-Itano et al., 1998; Carney and Moeller, 1998; Moeller, 2000). Moreover, when the same identification and intervention benchmarks are achieved (prior to 6 months of age), children perform as much as 20 to 40 percentile points higher on school-related measures (reading, arithmetic, vocabulary, articulation, intelligibility, social adjustment, and behavior) (Yoshinaga-Itano, 1995; 2003; Yoshinaga-Itano and Sedey, 2000). Therefore, early detection of hearing loss in infants and young children is justifiable as a priority in public health in general, with the specific responsibility placed on the health care subspecialty of audiology. 545

3 546 Section III Special Populations REFERRAL PATTERNS IN PEDIATRIC HEARING HEALTH CARE The early identification of hearing loss and follow-up diagnostics assessment are currently carried out through referrals from a variety of sources. These sources include statemandated newborn hearing screening programs and early intervention agencies, primary care providers who coordinate regular medical home visits, and other medical specialists (e.g., otolaryngologists, developmental pediatricians, neurologists) involved with infants and young children. This early referral has been advocated based on the mounting evidence that the earlier confirmation of hearing loss occurs, the earlier intervention can begin, thereby increasing the likelihood of optimizing a child s potential in all developmental areas. Early Hearing Detection and Intervention Program Referrals In recent years, the emphasis on universal early detection of hearing loss in infants has grown considerably. For example, in 1993, only 11 hospitals were screening more than 90% of their newborns. By 2005, every state had implemented a newborn hearing screening program, and data suggest that about 95% of newborn infants in the United States are screened for hearing loss prior to hospital discharge (National Center for Hearing Assessment and Management [NCHAM], 2007). During this same time period, it became clear that screening is only the first step in a process necessary to identify infants and young children with hearing loss. The next step is to provide these children and their families with timely access to diagnostic follow-up and, when necessary, referral to appropriate, culturally sensitive, and family-centered intervention services. It should be noted that all professional groups involved in the early detection of hearing loss have replaced the phrase universal newborn hearing screening (UNHS) programs with the term early hearing detection and intervention (EHDI) programs. This conceptual change was introduced when it became apparent that screening was just a critical first step. Experience has shown that, in order to successfully identify and serve infants and young children with hearing loss and help their families, professionals must go beyond screening. To be successful, all elements of follow-up need to be included to meet and serve the child s and family s complex needs. To successfully serve infants and young children with hearing loss, an agency within the EHDI program (usually the State Department of Health) has the responsibility of tracking and initiating referrals. To achieve optimal follow-up benchmarks, State Departments of Health and state Part C programs should be working closely together. Part C of the Individuals with Disabilities Education Act (IDEA) mandates responsibility for Child Find and intervention for children with disabilities, and requires all states to provide appropriate early intervention programs for all infants and young children meeting the disability criteria. As such, systematic communication is required between State Departments of Health and Part C programs to facilitate referrals to audiologists. This is achieved when a hospital s newborn hearing screening outcomes are communicated in parallel to State Departments of Health and state Part C programs. In turn, audiologic referrals can and should be made as soon as possible by Part C program coordinators. Clearly, well-coordinated tracking and surveillance systems must be developed by states to achieve the desired outcomes. Optimally, within the limits of confidentiality, each service provider within the EHDI system (e.g., hospital, practitioner, public health agency, public and private education agencies) participates in information management to track elements of care to each infant and family. The Child s Medical Home It is clear that services for infants and young children suspected of or at risk for hearing loss would be optimal if they were connected soon after birth to a primary care physician who is familiar with the children s circumstances, is knowledgeable about the consequences of hearing loss, is an advocate/facilitator for early referral, and is known and trusted by the family. To this end, primary care physicians, working in partnership with parents and other health care professionals including audiologists, make up the infant s medical home. A medical home isdefinedasanapproach to providing health care services where care is accessible, family-centered, continuous, comprehensive, coordinated, compassionate, and culturally competent. The primary care physician acts in partnership with parents in a medical home to identify and access services needed in developing a global plan of appropriate and necessary health and habilitative care for infants identified with hearing loss (American Academy of Pediatrics, 2002; Tonniges and Palfrey, 2004). Thus, the medical home serves as an important link between families, all service providers, and the state early intervention agency. As newborn hearing screening programs have reached the benchmark of screening a minimum of 95% of newborns, the goal of 95% of those referred achieving follow-up has been difficult. The State of Colorado reports that 76% of infants achieved follow-up audiologic evaluations to confirm or exclude congenital hearing loss (Mehl and Thomson, 1998). In a 1999 review of the Texas screening program, 97% of newborns were screened, yet only 64% returned for follow-up (Finitzo and Crumley, 1999). These numbers indicate the challenge of continuity of care from screening to audiologic follow-up and confirmation of hearing loss. As such, the medical home is part of a system responsible for assuring that a child is referred and scheduled for audiologic follow-up in a timely manner. The physician has the opportunity to articulate to parents the importance of

4 Chapter 23 Assessment of Hearing Loss in Children 547 FIGURE 23.1 The distribution of hearing loss etiology into genetic causes and environmental causes. audiologic follow-up and to actively demonstrate the importance by facilitating early referrals to audiologists. Moreover, the child s medical home ensures that children and families enter the health care and early intervention system with minimal obstacles and that they attend appointments. Other Medical Specialists The most effective approach to the early detection of and intervention for hearing loss involves a multidisciplinary team working individually and collectively to assess a patient. As such, early referrals to appropriate professionals are essential to facilitate accurate diagnosis and, in turn, result in efficient and effective patient care plans. A wide array of health care providers (e.g., primary care physicians, audiologists, otolaryngologists, ophthalmologists, geneticists, developmental pediatricians) and education service providers (e.g., speech-language pathologists, educators of children who are deaf or hard of hearing, early intervention professionals involved in delivering EHDI services) is essential given the complexity of hearing loss. Hearing loss can exist as a single entity, or it may be one aspect of multiple anomalies. Figure 23.1 details the distributions of etiologies relative to hearing loss, providing information on the biologic complexity of individuals with hearing loss. As such, early referrals between and among specific service providers are essential for a thorough assessment of the individual suspected of hearing loss. With the current focus on early detection of hearing loss, referrals to audiologists are occurring at earlier and earlier ages. The importance of audiologic diagnosis of hearing loss cannot be overstated; often the identification of hearing loss may be the first indication of a compromised sensory system. In turn, the medical workup of the patient may lead to additional findings of other sensory system or body system involvement. In concert with audiologic care coordination by the audiologist, every infant with a confirmed hearing loss must be evaluated by an otolaryngologist with pediatric expertise and have at least one exam by an ophthalmologist experienced in evaluating infants. Additionally, Figure 23.1 clearly suggests that families should be offered a genetics consultation. This evaluation can provide families with information on etiology of hearing loss, prognosis for progression, associated disorders (e.g., renal, vision, cardiac), and likelihood of recurrence in future offspring. Finally, the success of EHDI programs depends on families working in partnership with professionals as a wellcoordinated team. The roles and responsibilities of each team member should be well defined and clearly understood. ESTABLISHING THE ETIOLOGY OF HEARING LOSS As with all disorders, early detection enables early intervention and improves prognosis. In the case of hearing loss, early detection can influence the type of educational programming for the child and the need for family counseling. After a hearing loss is confirmed, at the same time as early intervention is begun, consideration should be given to identify the etiology of the hearing loss. As greater numbers of children with hearing loss are identified early through newborn hearing screening, more

5 548 Section III Special Populations evidence regarding the distribution of different etiologies may provide additional insights. However, long-standing data on profound hearing loss in infancy indicates that 50% of congenital hearing loss is thought to be due to environmental factors and 50% is hereditary (Fig. 23.1) (Gorlin et al., 1995; Morton, 1999). Hearing loss is a feature in over 600 syndromes (usually named after an individual or individuals [e.g., Waardenburg syndrome, Down syndrome] or the prominent features of the condition [e.g., brachio-oto-renal syndrome]) where the affected individual has other characteristics in addition to hearing loss (Nance, 2003). Conversely, nonsyndromic hearing loss is named according to the following scheme (DFN = deafness): DFNA: autosomal dominant disorders DFNB: autosomal recessive disorders DFN: X-linked disorders A genetic trait determined by its own pair of genes can be inherited in one of these three modes. It is the trait itself (and not the gene) that is dominant, recessive, or X-linked. However, for simplification, genes are often referred to as dominant, recessive, or X-linked. A dominant trait is expressed when one copy of the gene pair codes for the trait. In autosomal recessive genetic disorders, a person inherits two copies of an autosomal gene with a change (mutation) in both copies. In X-linked inheritance, genes are located on the X chromosome and can be recessive or dominant. Over 120 genes associated with hearing loss have been identified, and this number grows annually (Nance, 2003). In addition, remarkably, a single gene called GBJ2 that codes for a protein named connexin 26 gives rise to more than half of the genetic cases of hearing loss in the United States (Genetic Evaluation of Congenital Hearing Loss Expert Panel, 2002). As the widespread use of newly developed vaccines decreases the prevalence of etiologies such as measles, mumps, rubella, and childhood meningitis, the percentage of early-onset hearing loss attributable to genetic etiologies should increase. For these reasons and because family planning for subsequent children may be under consideration by the family, the multidisciplinary team, whether formally developed or informally constituted by the primary care physician in the medical home, should depend on the geneticist to establish the etiologic basis for hearing loss. AGE-APPROPRIATE ASSESSMENT The use of age-appropriate techniques in diagnostic audiology is vital in the evaluation of infants and young children. It requires clinicians to select differential diagnostic techniques that are within the child s developmental capabilities. Because children undergo rapid sensory, motor, and cognitive development and because some children will present with multiple developmental problems, it is vital that assessment tools are appropriate for the neurodevelopmental status of the young child. Factors (physical and cognitive) that can influence developmental status must be considered prior to the selection of an assessment strategy. Some problems may be relatively easy to identify (e.g., cerebral palsy). Others (e.g., learning disabilities, Asperger syndrome, or others on the autism spectrum) are more difficult to identify. Of course, knowledge of the handicapping conditions will enable the audiologist to plan effective diagnostic strategies. The audiologist must investigate the factors involved in each individual by use of interviews, case histories, assessments by other professionals, and close observation. Provision of appropriate services depends on a thorough knowledge of the individual to be served and his or her family background. COMPREHENSIVE AUDIOLOGIC ASSESSMENT The goal of the initial diagnostic assessment of infants and young children is to confirm or rule out hearing loss, to quantify the magnitude and configuration of hearing loss, and to assess the integrity of the auditory system. Additionally, comprehensive assessment should be provided for each ear even if only one ear was in question from the newborn hearing screening. A comprehensive audiologic assessment must be viewed as a process and not an isolated clinical visit. Serial evaluations may be necessary to develop reliable profiles of hearing status and developmental abilities. It is not uncommon for an audiologist to formulate a working diagnosis of the child s audiologic status in parallel with developing initial management options. Thus, ongoing assessment is an integral part of the management process. After the initial audiologic assessment is completed, frequent follow-up visits are essential in order to monitor infants overall auditory status, their development of auditory skills, and their functional use of hearing. Moreover, it must be recognized that single-point assessment does not adequately address the issue of fluctuating and/or progressive hearing loss. The Test Battery Approach The initial audiologic test battery to confirm hearing loss must include physiologic measures and, when developmentally appropriate, behavioral methods. The use of any test alone for assessing children s hearing sensitivity is discouraged. The desirability of using multiple tests in clinical practice is based on the complex nature of the auditory mechanism and the fact that auditory dysfunction may result from pathology at one or more levels of the auditory system. In test battery selection, the audiologist should use test procedures that are outcome based and cost effective, and greater weight should be given to the results of those tests for which validity and reliability are highest. If test results are not in agreement, the reason for the discrepancy must be explored before arriving at an audiologic diagnosis. Jerger and Hayes (1976) promoted the concept of a test battery approach so that a single test is not interpreted in

6 Chapter 23 Assessment of Hearing Loss in Children 549 isolation but, instead, various tests act as cross-checks of the final outcome. Hanley (1986) noted that cross-checks not only establish what the auditory disorder is, but also clearly establish what the auditory disorder is not. Thus, audiologists benefit by having a battery of tests appropriate for the diagnosis of hearing loss in infants and young children. As pointed out by Turner (2003), the purpose of multiple tests is to increase the accuracy of audiologic diagnosis. This is accomplished when appropriate diagnostic tests are selected for the individual s test battery. Subsequently, tests must be carefully administered and data appropriately interpreted, followed by a clinical decision based on the entire test battery. After weighing the agreement/disagreement between tests, the audiologist can reach a confident diagnosis. Clinical decision involves not only test selection, but also determining the number of tests administered during a single session, interpreting individual test data, and then drawing conclusions based on the performance of the entire test battery. PEDIATRIC AUDIOLOGIC ASSESSMENT PROCEDURES Audiologic assessment of infants and young children includes a thorough case history, otoscopic inspection, and both physiologic and behavioral measures. As stated earlier, the need for a battery of tests in pediatric assessment is essential in order to optimally plan for and meet the diverse needs of the pediatric population (e.g., age, physical status, developmental level, neuromaturational level). Case History The case history is a component of the audiologic assessment that guides the audiologist in constructing an initial developmental profile based on the child s physical, developmental, and behavioral performance. It also can serve as the first cross-check on the audiologic test outcome. Some background health-related and developmental information (American Speech-Language-Hearing Association [ASHA], 2004) can be obtained prior to the initial evaluation (by telephone and/or mail). However, the majority of information is generally obtained at the time of the evaluation in a face-to-face interaction with the child and family. Face-to-face interaction provides the opportunity for posing questions based on observations, initial impressions, and interactions with the child. Moreover, this is an excellent time to establish rapport with the child and parent(s)/caregivers. The outcome of the case history is particularly important because it will often guide the strategy for the audiologic assessment and for making subsequent recommendations and referrals. Otoscopic Inspection Otoscopy is intended as a general inspection of the external ear and tympanic membrane for obvious signs of disease, malformations, or blockage from atresia, stenosis, foreign bodies, cerumen, or other debris. Moreover, because several audiologic assessment procedures require the insertion of a probe into the external auditory canal, the visual inspection serves to verify that there is no contraindication to placing a probe in the ear canal. AUDIOLOGIC TEST BATTERY: BIRTHTO6MONTHSOFAGE Ear-specific assessment is the goal for both behavioral and physiologic procedures because a unilateral hearing loss, even in the presence of one normal-hearing ear, may place a child at significant developmental and/or educational risk (Bess, 1982; Bess et al., 1988; 1998; Bovo et al., 1988; Oyler et al., 1988). Therefore, determining hearing sensitivity for each ear facilitates medical/surgical diagnosis and treatment, selecting and fitting amplification when appropriate, establishing baseline function, and monitoring auditory status when progressive, fluctuating, or late-onset hearing loss is suspected. To be developmentally appropriate, the audiologic test battery for young infants, birth through 6 months, consists primarily of physiologic measures. These measures currently include the auditory brainstem response (ABR) as the gold standard, the auditory steady-state response (ASSR) to supplement/augment ABR findings, distortion product or transient-evoked otoacoustic emissions, and acoustic immittance. (For a more complete discussion of these procedures, see Chapters 14 and 15.) Measurement of auditory evoked potentials, especially the ABR, can provide accurate estimates of threshold sensitivity in young infants. To maximize reliable electrophysiologic measurements in infants, an adequate signal-to-noise ratio (S/N) must be maintained, and an extended recording window must be used to identify threshold responses. The number of signals averaged may vary according to the amount of background noise, the response amplitude, and the presence of hearing loss. In addition, for stimulus conditions close to threshold and for frequency-specific stimuli, especially low-frequency stimuli, a 20- to 25-ms recording window is essential. The ABR protocol with infants should include frequency-specific stimuli using insert earphones, unless contraindicated, for air-conduction testing. For these measurements, short-duration, rapid-onset tone bursts are used. Knowledge of the spectra of these stimuli is needed because they are affected by various stimulus parameters. When airconduction thresholds obtained by ABR are found to be abnormal, estimates of bone-conduction sensitivity should be completed as well. Bone-conduction ABR is important for quantifying the degree of the conductive component when hearing loss is present. It is important when doing boneconduction ABRs that attention is paid to ensure adequate pressure of the bone vibrator on the mastoid (Yang and Stewart, 1990). Additionally, because of stimulus artifact

7 550 Section III Special Populations concerns, care must be taken to separate the bone vibrator from the electrode due to electromagnetic leakage. At a minimum, responses to low- and high-frequency stimuli should be obtained for each ear to estimate audiometric configuration. High-frequency assessment should use 2,000-Hz tone bursts (The Pediatric Working Group, 1996), and the low-frequency assessment should use 250- or 500-Hz tone bursts (Stapells et al., 1995; Stapells and Oates, 1997). These data not only strengthen behavioral estimates of hearing loss, but they also facilitate the selection and fitting of amplification (hearing aids and frequency modulation [FM] systems). Hearing aid fitting protocols for use with infants that are based on frequency-specific information are now available (American Academy of Audiology, 2003; The Pediatric Working Group, 1996; Seewald et al., 1997; Stelmachowicz, 2000). Although the use of click stimuli alone is not sufficient for the estimation of audiometric configuration, a click stimulus can provide useful information regarding neural integrity. Assessment of interwave latencies, ear asymmetries, and morphology relative to age-appropriate norms may be completed as part of the ABR assessment and used along with other clinical and/or medical findings. If there are risk indicators for neural hearing loss (auditory neuropathy/auditory dyssynchrony [AN/AD]) such as hyperbilirubinemia or anoxia (although some AN/AD patients have no risk factors and are found in the well-baby nursery [e.g., nonsyndromic recessive AN/AD such as that related to otoferlin-based AN/AD]), then audiologic assessment should include click-evoked ABR. When recording a high-level (80 to 90 db normal hearing level [nhl]) click ABR, responses should be measured separately for condensation and rarefaction single-polarity stimuli, and responses should be displayed in such a way as to identify the cochlear microphonic (CM) (i.e., superimposing averages to identify out-of-phase components). In these instances, precautions must be taken to distinguish the CM from a stimulus artifact. Many children in this age group can be tested during natural sleep, without sedation, using sleep deprivation with nap and feeding times coordinated around the test session. However, active or older infants may require sedation to allow adequate time for acquisition of high-quality recordings and sufficient frequency-specific information. The ASSR is a synchronized brain response to modulated tones with emerging clinical applications. The ASSR uses a continuous frequency-specific stimulus that is modulated (i.e., frequency and/or amplitude modulated) and presented at a given frequency. The recorded response is generated in the electroencephalogram (EEG) response rather than specifically in the auditory brainstem pathway, as is the case with the ABR. Whereas the ABR response is determined through the identification of peaks and troughs in the time domain, the presence or absence of the ASSR is determined through statistical algorithms in the frequency domain. Clearly, the ASSR holds promise as a method of estimating frequency-specific hearing sensitivity in individuals; however, more data are needed regarding the predictive accuracy of the ASSR as a function of age, stimulus type, recording time, and the magnitude of hearing loss. Because of its recent development, the ASSR does not have the evidence-based underpinnings with diverse clinical populations to recommend it as the sole measure of auditory status in the infant and young child populations. For example, issues related to threshold estimation within the first three months of life have been reported (John, et al., (2004) and some concerns about recording artifact under certain stimulus conditions have been expressed (Small andstapells, 2004; Gorga, et al., 2004). Research in this area is ongoing, and improvements/recommendations in methodology are expected. Otoacoustic emissions (OAEs) expand the pediatric audiology test battery by providing a physiologic means of assessing preneural auditory function (Gorga et al., 1993; Kemp et al., 1990; Norton and Widen, 1990). That is, OAEs are most likely generated by the outer hair cells in the cochlea and serve as an indirect measure of these hair cells. OAEs are not, in and of themselves, necessary for hearing, nor are they a mechanism of hearing, but rather, they reflect the status of structures that are necessary for hearing. Evoked OAEs occur in response to an external auditory stimulus and are present in nearly all normal-hearing individuals. Thus, the presence of OAEs is consistent with normal or near-normal hearing thresholds in a given frequency region. Moreover, measuring OAEs clinically permits the differentiation between sensory and neural components of the sensory-neural hearing loss (Lonsbury-Martin et al., 1993). Used in conjunction with ABR, OAEs are useful not only in the differential diagnosis of cochlear hearing loss but also in the identification of infants and young children with neurologic dysfunction. Transient-evoked OAEs (TEOAEs) are elicited following a transient (click) stimulus at approximately 80 db peak sound pressure level (SPL). Although the transient click stimulus is a broadband stimulus that is not frequency specific, the response is analyzed in the frequency domain, thus providing information across frequencies from 500 to 5,000 Hz, although test performance is best for mid to high frequencies. Distortion product OAEs (DPOAEs) are elicited following stimulation with two tones. DPOAEs are measured in response to two tones (primaries) that interact to produce nonlinear distortions in the cochlea. The two tones typically are selected so that the frequency ratio between the tones (f2/f1) is 1.22, which is known to produce the largest distortion product at most test frequencies in humans. Response criteria typically include S/N and/or have a response reproducibility of greater than an established percentage at defined frequencies (see Chapter 21 for further details). Schemes for trying to determine the degree of hearing loss and/or predicting thresholds using OAEs have been investigated (Martin et al., 1990; Boege and Janssen, 2002; Dorn et al., 2001; Gorga et al., 1996; 2002; 2003). Although

8 Chapter 23 Assessment of Hearing Loss in Children 551 some strategies have met with success, variability is such that threshold predictions should be viewed cautiously. Because of their remarkable stability over time within the same ear, OAEs also are useful for monitoring the status of disease conditions that are progressive, including certain genetic disorders such as Usher syndrome (Meredith et al., 1992). In addition, over shorter time courses, OAEs are advantageous for monitoring the effects of treatments that are potentially damaging to the ear, like those involving such ototoxic antibiotics as tobramycin (Katbamna et al., 1999) or such antitumor agents as cisplatin (Ress et al., 1999). Acoustic immittance measures are an integral part of the pediatric assessment battery. Clinical decisions should be made on a quantitative assessment of the tympanogram, including consideration of equivalent ear canal volume, peak compensated static acoustic admittance, tympanometric width or gradient, and tympanometric peak pressure (see Chapters 8 to 10 for a detailed description of the components of the acoustic immittance test battery). Under the age of approximately 4 months, interpretation of tympanograms may be compromised when a conventional low-frequency (220- or 226-Hz) probe tone is used (Paradise et al., 1976; Purdy and Williams, 2000). As such, a higher probe-tone frequency (e.g., 1,000 Hz) is recommended for identifying middle ear disorders in infants less than 4 months of age, and normative data for 1,000-Hz tympanometry are now available for neonates and young infants (Margolis et al., 2003). Once a child reaches the age of 7 months, a low-frequency (226-Hz) probe tone is appropriate. Between 5 and 7 months of age, however, there is still a possibility of false-negative tympanograms in ears with middle ear effusion. Therefore, use of a 1,000-Hz probe tone for tympanometry in this subset of infants is recommended when attempting to identify middle ear effusion. When a quantitative assessment of a tympanogram is used, care must be taken to ensure that there is correspondence between the graphic representation of the tympanogram and the absolute quantities indicated. With the pediatric population, sometimes there are irregularities in the tympanogram shape (due to movement artifact, crying, or vocalizing) that may be mistaken for a tympanogram peak by the instrument and may provide misleading absolute values. In addition to providing confirmation of middle ear status, acoustic reflex measurement is useful in the interpretation of other components in the audiologic test battery. That is, the acoustic reflex may provide supplemental information relevant to the functional status of the middle ear, cochlea, and brainstem pathway (see Chapters 10 and 22). For example, acoustic reflexes are absent when AN/AD exists (Starr et al., 1996). Although there are insufficient data for routine use of acoustic reflex measurements in the initial diagnostic assessment under the age of 4 months, the acoustic reflex should be used to supplement the test battery at older ages. Together, these measures are fundamental components of the pediatric audiology test battery. AUDIOLOGIC TEST BATTERY: INFANTS6MONTHSOFAGE AND OLDER For newborns and young infants (<6 months), a physiologic measure is the approach of choice when attempting to define an individual s auditory sensitivity. However, as valuable as physiologic procedures are in the early confirmation of hearing loss, the audiologist inevitably returns to behavioral testing to substantiate test results and monitor a child s hearing longitudinally. Therefore, assessing auditory sensitivity in older infants and children (>6 months) can be completed efficiently and effectively with both behavioral and physiologic measures. The audiologic test battery for infants age 6 months of age and older includes conditioned behavioral audiometry (either visual reinforcement audiometry [VRA] or conditioned play audiometry [CPA]), OAEs, acoustic immittance, and speech detection and/or recognition measures. ABR should be performed, as necessary, when behavioral measures are not sufficiently reliable to provide ear-specific estimates of type, degree, and configuration of hearing loss or when additional physiologic data are necessary to support other clinical questions (e.g., neurologic status). Moreover, the desire for behavioral hearing test results should not delay the selection and fitting of amplification when valid and reliable frequency-specific threshold information is available by physiologic measurement. Behavioral Observation Audiometry It is now known that unconditioned behavioral observation techniques with infants and young children are confounded by poor test-retest reliability and high inter- and intrasubject variability (Weber, 1969; Thompson and Weber, 1974). This places limitations on the use of behavioral observation audiometry (BOA) for determining auditory sensitivity. Therefore, BOA is no longer recommended for assessing frequency-specific threshold sensitivity in newborns, young infants (<5 months), or those children whose developmental disabilities preclude them from learning operant conditioning procedures. However, another goal in pediatric assessment is to examine auditory function. Although ABR can quantify auditory sensitivity in infants with compromised cognitive function, BOA can provide useful insight into the quality of the child s auditory responsiveness. Moreover, BOA can provide an estimate of functional capabilities useful in planning intervention for these children. That is, the audiologist can predict potential difficulties in auditory development and recommend aural habilitation strategies intended to improve the child s functional use of sound. PROCEDURAL GUIDELINES Infants under a developmental age of 5 months generally display a variety of reflexive and orienting responses to external

9 552 Section III Special Populations stimuli. The accurate judgment of these behaviors, however, may require the use of multiple examiners. One examiner can present auditory stimuli, while one or more examiners monitor state changes and cues signal presentations when the child s listening state is appropriate. Multiple examiners also may be necessary to judge response behaviors in order to reduce two common errors of observation: (1) judging that a response occurred when, in reality, there was no response and (2) judging that no response occurred when, in reality, a response did occur. Because so many behaviors are monitored (e.g., head or limb reflex, increased motion, decreased motion, wholebody startle, eye widening, nonnutritive sucking, searching, eye blink or flutter, localization, smiling, laughing, pointing), it is important to minimize false-positive responses of the observers. To minimize the examiner s expectations of the outcome of the test, the observer(s) should not be informed about the child s developmental status, previous test results, or medical status. Therefore, examiners should have minimal information about each patient before testing. Response behaviors seen during BOA can be separated into those that are attentive-type, orienting behaviors (e.g., increased and decreased motion, eye widening, searching, localization, smiling, laughing, pointing) and those considered reflexive (e.g., head or limb reflex, whole-body startle, sucking, eye blink or flutter). Analyzing response behaviors may provide useful information in determining how youngsters attach meaning to sound. If response behavior is reflexive from a child known to have near-normal hearing, it may indicate a child who does not attach much functional meaning to sound. On the other hand, children who show orienting responses to sound are likely demonstrating a higher level of cortical functioning. Operant Conditioning It is possible to approach the assessment of infants and young children behaviorally through operant conditioning paradigms, specifically through an operant discrimination procedure. In an operant discrimination procedure, a stimulus is used to cue the child that a response results in reinforcement. That is, the stimulus serves as a cue to perform a specified behavior (i.e., a behavioral response). In turn, operant behavior (the response) is increased by the application of positive reinforcement. Reinforcement is used to strengthen an easily monitored single response and keep the child in an aroused and motivated state. If the reinforcement is sufficiently powerful, the response will be continued over repeated presentations of the stimulus. As demonstrated by Moore et al. (1975; 1977), audiometric signals (e.g., puretones or warble tones) have limited reinforcing properties. Consequently, a positive reinforcement having high interest value (appealing to the infant) must be used to reinforce the response behavior. Maintaining motivation and a high response probability through the use of age-appropriate reinforcement allows audiologists to investigate, over time, an infant s or young child s auditory response behavior. This allows more precise estimation of ability (i.e., hearing sensitivity) and reduces the habituation found in behavioral assessment without reinforcement (i.e., BOA). Conditioned behavioral audiometry is an efficient and cost-effective approach for clinical use. However, this requires knowledge of potential pitfalls, as well as use of procedural modifications/enhancements that increase reliability and validity (Diefendorf and Gravel, 1996; Renshaw and Diefendorf, 1998). For example, potential pitfalls include the overactive child, controlling and minimizing false responses, and the timeliness and proper use of reinforcement. As such, conditioned behavioral audiometry requires knowledge of operant conditioning, the proper use of distraction, the proper use of reinforcement, and the importance of developmental age on successful outcomes. Beyond the time, cost, and ease of administration, the most important advantage of behavioral tests is that they allow infants and young children to demonstrate actively what they perceive. In turn, this fosters a valid description of their functional hearing abilities. Visual Reinforcement Audiometry Normally developing infants make head turns toward a sound source in the first few months of life. This neuromotor response undergirds the behavioral approach audiologists use to investigate auditory behavior. Indeed, by the time an infant has reached a chronologic/developmental age of 5 to 6 months, operant conditioning coupled with this behavioral response enables audiologists to implement VRA. The success of VRA is related to the fact that the response (a head turn; Fig. 23.2) and reinforcer (three-dimensional visual animation that is seen when the head is turned toward it) are well suited to the developmental level of children FIGURE 23.2 A head turn response coupled with visual reinforcement (a three-dimensional, mechanical toy housed in a smoked Plexiglas enclosure).

10 Chapter 23 Assessment of Hearing Loss in Children 553 between 6 months and 2 1 / 2 years of age. Once the child is under stimulus control, he or she will continue to respond at low sensation levels long enough to provide an accurate estimate of threshold. In VRA, conditioned head turns are reinforced by an attractive three-dimensional animated toy that is activated near the source of the sound that is presented. Visual stimuli containing movement, color, and contour appear to be more effective reinforcement than less complex visual stimuli. Complex visual reinforcement, such as three-dimensional animated toys, is critical for maintaining response behaviors over repeated trials (Moore et al., 1975). The visual reinforcement is frequently housed in a smoked Plexiglas enclosure. Activation of the visual reinforcement results in lighting the enclosure and animation of the toy (Fig. 23.2). The success of VRA also is related to the developmental status of the child being examined. Developmental issues that must be recognized include the impact of corrected age (corrected age is determined by subtracting the estimated weeks of prematurity from the infant s chronologic age) on VRA performance and the impact of mental age on VRA performance. Moore et al. (1992) concluded that VRA performance is related to corrected age. They studied 60 premature infants (36 weeks of gestation or less) at corrected ages of 4 to 9 months. Their results imply that premature infants with a corrected age of 8 or 9 months are likely to perform acceptably in response to VRA (can be conditioned and respond with high success before habituation to task); that premature infants with a corrected age of 6 or 7 months may perform but with less success (can be conditioned but have limited responses before habituation to the task); and that premature infants with a corrected age of 4 or 5 months are not likely to respond to the VRA procedure. A comparison of these data to results of previous studies on full-term infants demonstrates that although full-term infants are likely to respond with high clinical success to VRA by a chronologic age of 6 months (Moore et al., 1977), premature infants are not likely to respond to VRA with good clinical success until approximately a corrected age of 8 months. Widen (1990) evaluated VRA as a function of developmental age in premature, high-risk babies. Clearly, the developmentally mature babies were more often tested successfully (the ability to be conditioned and provide threshold for at least one stimulus). The data from Moore et al. (1992) and Widen (1990) are highly consistent; that is, both reports indicate that VRA success with premature infants is related to corrected age and that VRA success with these infants is greater as they approach 8 to 9 months of corrected age. Why is it that premature infants, even after prematurity is corrected for, lag several months behind normally developing, full-term infants? Premature infants have been shown to display significantly poorer performance on standardized measures of mental ability (Bayley Scales of Infant Development [BSIDs]) when compared to full-term infants of the same postpartum age (Rubin et al., 1973; Goldstein et al., 1976). Moreover, Kopp (1974) concluded that the quality and quantity of cognitive exploration, which preterm infants engage in less when compared with full-term infants, also may account for reduced motor development. Several studies have reported on the use of VRA in children with Down syndrome and other developmental disabilities. Greenberg et al. (1978) reported on the use of VRA in 46 individuals with Down syndrome between the ages of 6 months and 6 years. As would be expected, the proportion of successful tests increased as age increased. Because it is expected that chronologic age would be a very poor predictor of success with the VRA procedure in these children, the BSIDs (Bayley, 1969) were used to provide an estimate of developmental level. If children with Down syndrome are considered on the basis of their developmental age, in contrast to chronologic age, it would be logical to assume that results might be similar to those found with normally developing infants. However, whereas Wilson et al. (1976) found that normally developing infants 6 months of age and older accomplished the VRA procedure with a high rate of success, Greenberg et al. (1978) found that individuals with Down syndrome did not achieve a high rate until 10 to 12 months BSID mental age equivalent. These investigators further pointed out that when one is predicting potential success with the VRA procedure for children with Down syndrome, the BSID mental age equivalent score provides the most distinct distribution between successful and unsuccessful tests, with the dividing point being a BSID mental age equivalent of at least 10 months. Similarly, Wilson et al. (1983) reported that 80% of the children with Down syndrome, in their study, were testable by 12 months of age using VRA. For the child with special needs, Thompson et al. (1979) indicated that VRA was an effective test procedure for 88% of the low-functioning children in their study, if all children under a developmental age of 9 months were excluded. One challenge facing the audiologist using VRA for the child with special needs is that the child may often be older than ideal for this type of operant procedure. Gravel and Traquina (1992) noted that younger infants (6 to 18 months) were easier to assess than older toddlers (18 to 24 months), and children ages 21 to 24 months were less tolerant of earphones, more distractible, and less interested in the reinforcers. In summary, when VRA is considered in clinical protocols (Fig. 23.3), audiologists must consider: (1) corrected age adjusted for prematurity rather than chronologic age or (2) mental age/developmental age when disparities exist between corrected age and the child s developmental status. Of the two predictors of VRA performance (corrected age and mental age), corrected age may be the more practical one to use in most cases because it can be obtained from parental report, case history information, and/or hospital records and does not require tests necessary to determine mental age. TEST ROOM ARRANGEMENT Figure 23.4 presents the room arrangement most commonly used for VRA. The audiologist in the control room has full

11 554 Section III Special Populations FIGURE 23.3 Age considerations for optimizing success in visual reinforcement audiometry (VRA). FIGURE 23.5 A second examiner maintaining the interest of an infant in a midline position. FIGURE 23.4 Test room arrangement commonly used in visual reinforcement audiometry (VRA). view of the testing situation. The ability to selectively darken only the audiologist s side of the test booth can be helpful. The infant, parent, and a second examiner are located within the test suite. The second examiner, seated at the infant s side, maintains the infant s head in a midline position by quietly encouraging the child to observe passively or to play with colorful, nonnoisy toys (Fig. 23.5). An appealing toy is manipulated by the examiner in front of the infant as a distractor. The examiner s role is to maintain the infant s attention at midline and return the infant to this position once a response is made and reinforcement is completed. The audiologist must be creative in keeping the child alert and in a listening posture without the child becoming so focused on the activity. The toys used for this purpose should be appealing but not so attractive as to overly occupy the infant s attention. The closer the audiologist is to the child, the more easily the child is engaged in the activity. If the infant under test shows too much interest in the colorful, nonnoisy toys, then the potential exists for no response during a signal trial or for elevated response levels due to decreased attention. Conversely, if the examiner and toys are not sufficiently interesting, the likelihood of false alarms (random looking toward the visual reinforcer) will be high. Often a touch of the hand on a child s shoe or leg will quietly redirect the child to a midline position. Actually sitting on the floor in front of the child allows for a totally unobstructed view of the child for the control room audiologist and being located slightly below the level of the child is a nonthreatening position. Obviously, it is difficult to balance entertaining the child without totally consuming the child s attention. A single examiner in the control room can use a mechanical centering toy positioned at midline in the test room for maintaining a child s midline distraction. While the use of a centering toy is an alternative to the more traditional approach of using two examiners, this approach distances

12 Chapter 23 Assessment of Hearing Loss in Children 555 the single examiner (in the control room) from the infant (in the exam room). Although a single examiner may be more cost effective and clinically practical in busy settings, it also reduces the flexibility of maintaining the infant at midline with a second examiner and multiple distraction toys. That is, the centering toy is always the same toy, is somewhat noisy, and may be too much distraction for some infants. Additionally, the centering toy may compete with the actual visual reinforcers because they are so much alike. The fact that the centering toy is not housed in a dark Plexiglas box raises the concern that the constant viewing of the animated toy may result in less interest with the visual reinforcers. When only one examiner is used for assessment, the use of computer-assisted test procedures can overcome some of the disadvantages described above. That is, a microcomputer interfaced with a clinical audiometer can be controlled from the test room, allowing the examiner to maintain a closer position to the infant in the exam room. Depending on the child s acceptance and responses, insert earphones or headphones can be used, rather than the sound field speakers, to obtain ear-specific information during VRA testing. The preferable transducers are insert earphones. Inserts are useful for a number of reasons, including their comfort and light weight, their increased interaural attenuation compared to conventional earphones, and the reduced risk of ear canal collapse. CONDITIONING IN VISUAL REINFORCEMENT AUDIOMETRY In clinical assessment, the first phase of VRA is the conditioning process. Response shaping is critical to the success of the operant procedure. Two different approaches that can be attempted in the first phase are (1) pairing a supra-threshold auditory stimulus with the visual reinforcer or (2) presenting a supra-threshold auditory stimulus and observing a spontaneous response from the infant, followed by activation of the reinforcer. Evidence suggests that different signals (e.g., tones, filtered noise, speech) are equally effective during the conditioning phase (Thompson and Folsom, 1984; Primus and Thompson, 1985). Successful completion of the training phase is the achievement of a pre-established criterion of consecutive head turn responses. If the criterion is not reached (usually two or three responses following two or three conditioning trials), phase 1 retraining is necessary until the criterion is met. The number of training trials needed before phase 2 trials begin varies, but the training phase is usually brief. A key to response shaping is the presentation of a suprathreshold stimulus. For most infants, supra-threshold will be 30, 50, or 70 db. However, some children, particularly those children with moderately severe to severe hearing loss, may require 90 db or higher to qualify as a supra-threshold stimulus. Because hearing status is unknown, supra-threshold estimates also are unknown. Therefore, the possibility exists that the stimulus selected to shape the response behavior might be inaudible. Given that most infants can be expected to have normal hearing, the most efficient test is one that uses a low starting level, approximately 30 db. However, failure to condition rapidly should alert the audiologist to a potential equipment/calibration problem or a child who requires a greater starting intensity for conditioning. Attention to either issue must be immediate for a successful outcome. A further clinical variation of this basic procedure is used for situations when the infant s head turn is not being shaped by the auditory stimulus at high-intensity levels. A bone oscillator is placed on the infant s mastoid on the side of the reinforcer. If necessary, the bone oscillator can be removed from the headband and held in the child s hand or rested against the child s arm to use as a vibrotactile stimulus. The traditional conditioning procedure is initiated, that is, pairing the stimulus with the reinforcement. The stimulus usually selected for bone-conducted conditioning is a 250-Hz narrowband noise presented at 60 db hearing level (HL). An infant with severe or profound hearing loss with no other developmental disabilities will show appropriate behavioral responses as long as the stimulus is salient (can be felt, even if not heard). Responses are obtained in this manner; subsequently, earphone or insert phone presentations follow with starting intensity levels dependent on the bone-conduction responses. If the youngster under test fails to display conditioned responding, other issues such as compromised physical status, developmental delay, or immaturity are raised. For example, Condon (1991) noted several cognitive attainments that are necessary for a child to be assessed reliably using VRA. The child must be developing object permanence (i.e., knowing that objects exist in space and time, even when the child can no longer see them or act on them) and the ability to anticipate the reappearance of at least partially hidden objects, discover simple causality (i.e., an event or behavior is dependent on the other for its occurrence) and means-end relationships (i.e., behaviors that result in anticipated outcomes), and use simple schemes to explore toys. Successful completion of training occurs when the infant is making appropriate responses and random head turning is at a minimum. Subsequently, the test phase of VRA begins. Signal intensity is attenuated after every yes response or increased after every no response. Using a conventional staircase (up-down) procedure, signal intensity is raised and lowered by 10 db. Testing is continued until a stopping criterion (ascending and descending until four reversals points have been achieved; refer to Fig. 23.6) is met. Threshold (minimum response level) is then defined as the mean of the reversal points. PROCEDURAL GUIDELINES The key to implementing valid and reliable behavioral techniques is to use procedures supported by clinical research. That is, many aspects of VRA have been reported in the literature that, taken together, provide the guidelines for clinical approaches.

13 556 Section III Special Populations FIGURE 23.6 Algorithm commonly used in visual reinforcement audiometry (VRA). The recommended trial duration (incorporating the signal and response interval) is approximately 4 seconds (Primus, 1992). That is, the signal duration is approximately 4 seconds, and this 4-second interval also defines the period of time during which a response is judged to be present or not. Responses outside of the 4-second period are not interpreted as valid responses. Visual reinforcement is provided only for correct responses that occur during signal trials. It is sometimes helpful to use the light initially, eventually adding the light plus animation. The novelty of the task may be preserved longer by introducing the more complex reinforcement as the testing progresses. Moreover, by starting with the light only, the audiologist can gauge any potential for a fearful response to the reinforcement. The novelty of VRA also may be strengthened with older children by using moving images generated by a digital video disc (DVD) player/monitor. Schmida et al. (2003) used digital video with 19- to 24-month-old children. Their results demonstrated a greater number of head turn responses before habituation when viewing video reinforcement than when viewing conventional animated toy reinforcement. These results support the hypothesis that the complex and dynamic nature of the video reinforcement would be more effective in achieving a greater number of responses than the conventional toy reinforcer prior to habituation in the 2-year-old age group. In general, a 100% reinforcement schedule (reinforcement for every correct response) results in more rapid conditioning, yet more rapid habituation. Conversely, an intermittent reinforcement schedule produces slower conditioning but also a slower rate of habituation. Consequently, most clinicians recommend a protocol that begins with a 100% reinforcement schedule and then gradually shifts to an intermittent reinforcement schedule. Primus and Thompson (1985) compared a 100% reinforcement schedule to an intermittent reinforcement schedule with 2-year-old children. The two reinforcement schedules resulted in no differences in the infants rate of habituation or the number of infant responses to stimulus trials. These findings provide an excellent guideline for delivering reinforcement. Since Primus and Thompson s data suggest that withholding reinforcement should not affect the amount of response behavior, reinforcement should not be provided if the audiologist is at all uncertain about the validity of an infant s head turn response (Diefendorf and Gravel, 1996). The risk of reinforcing a random head turn is that it may lead to confusion for a child during the test session and increase the child s rate of false responding. Failure to reinforce a correct head turn, however, does not degrade performance. In this situation, withholding reinforcement for a correct but ambiguous response is viewed as intermittent reinforcement, which will not interfere with subsequent infant behavior. Reinforcement duration also is a factor influencing response outcome from children around the age of 2 years (Culpepper and Thompson, 1994). Decreasing the duration of a child s exposure to the visual reinforcer (e.g., 4 seconds to 0.5 second) results in an increase in response behavior and a decrease in habituation. Audiologists may increase the amount of audiometric information obtained from children

14 Chapter 23 Assessment of Hearing Loss in Children 557 by decreasing their exposure to the visual reinforcer. For the child with special needs who may have a slower response, the visual reinforcer should be activated for a sufficient length of time for the child to very briefly observe it, but prolonged visual reinforcement should be avoided. Ensuring valid VRA outcomes depends on separating true responses from false responses (false positives) during threshold acquisition. Based on normal response behavior from adults, the assumption is made that infants also produce a number of incorrect responses during the testing phase of VRA. Throughout testing, two types of trials are presented: signal trials that contain a stimulus and control trials that do not. Reinforcement is provided only for correct responses during signal trials. False-positive responses are monitored by inserting control trials in the staircase algorithm (see Fig. 23.6). A response observed during a control trial (and never reinforced) is evidence of false responding. Thus, it is possible to systematically estimate errors or chance responding (false responses during signal trials) by calculating the number of responses during control trials. Moore (1995) recommended that one out of four presentations should be a control trial and that test results are questionable if the false-positive rate exceeds 25%. Eilers et al. (1991a) suggest that a false alarm rate of 30% to 40% is acceptable and adopting such a rate as acceptable does not compromise the accuracy of thresholds for clinical assessment. Clearly, high false alarm rates (>50%) require the audiologist to further consider that test results may be inaccurate. Excessive false responses suggest that the infant is not under stimulus control. As such, audiologists should focus on two factors to rectify clinical outcomes: (1) reinstituting phase 1 shaping and conditioning, and (2) increasing the entertainment level of the activity to engage the child s interest at a midline position before starting a test trial. When in the test room, an examiner must be able to choose from a variety of toys available and judge when a toy change in either direction (enhanced novelty and thus more entertaining, or simpler/less novel and thus less entertaining) is necessary to maintain the child s midline focus and optimum response state. Occasionally, overactive parents can bias their children to respond, thereby resulting in excessive false responses. Therefore, parents may need to wear headphones through which masking music or noise is delivered. Threshold determination in audiometry is based on the lowest intensity level where responses are obtained approximately 50% of the time. In VRA, as the staircase algorithm proceeds, how many reversals should be required before identifying the hearing threshold? Too few may sacrifice response accuracy. However, too many will increase test time, in turn reducing the number of stimulus presentations that could be spent obtaining thresholds to other stimuli. Testing of one stimulus may be stopped once the infant has exhibited between three and four response reversals (Eilers et al., 1991a; 1991b). Eilers and her colleagues found that using six rather than three response reversals before discontinuing the threshold search had minimal effect on threshold. Yet tests with a three-reversal stopping rule were significantly shorter than those with six reversals. As stopping rules are increased from three to six, there is about a 50% increase in the number of test trials, with no improvement in response accuracy. These results suggest that, by using relatively few reversals to estimate threshold, a staircase algorithm may be shortened to increase efficiency without sacrificing accuracy. Thus, there is no need to continue testing beyond three or four reversals since the results obtained are not substantially better because of it. Audiologists must proceed as if the next piece of information from the youngster under test will be their last. This strategy dictates air versus bone conduction, another frequency versus speech, or switching ears. If the second ear is tested and the child turns in the wrong direction, the reinforcer display is activated regardless. The light and animation from the reinforcer will attract the child s attention back to the reinforcer. Obviously, two reinforcers, one on each side of a test room, alleviate this concern, and although desirable, they are not necessary for successful assessment outcomes. Thresholds obtained with the VRA procedure for infants 6 to 12 months of age have been shown to be within 10 to 15 db of those obtained from older children and adults (Gravel and Wallace, 1998; Nozza and Wilson, 1984). In addition, VRA thresholds are similar across the age span (6 to 24 months) and show good reliability when compared to thresholds obtained from the same child at older ages (Diefendorf, 1988). CONDITIONED PLAY AUDIOMETRY Operant conditioning of behavioral responses to sound continues to be an effective approach for older children. What changes as children age, however, are the response behavior and the reinforcement that is used. Like the operant headturn procedure, play audiometry uses positive reinforcement to support response behavior. In CPA, children learn to engage in an activity (e.g., putting rings on a spindle, dropping or stacking blocks, putting together simple puzzles) each time they hear the test signal. These activities are assumed to be interesting to children, are within their motor capability, and represent a specific behavior that is used to denote a response to a stimulus. For youngsters with limited gross motor/fine motor skills, a variety of responses (e.g., finger swing, hand motion, arm motion, eye motion, visual gaze) can be used to trigger an electronic switch, in turn activating a computer screen programmed for appropriate visual reinforcement. The goal is to select the most appropriate task and the most interesting reinforcement while at the same time recognizing the physical limitations that may compromise the child s success. If the physical demands are too great, then the task will detract from maintaining a listening posture. If the task is too

15 558 Section III Special Populations simple, the child will have less motivation to participate and will tire of the task. The critical decision for the audiologist is to select a specific behavior that is used to denote a specific response to a stimulus. CPA follows the traditional operant conditioning paradigm of stimulus response reinforcement, in which the play activity/motor activity is the response behavior and social praise/another reward is the reinforcement. Three challenges exist in play audiometry that require a skillful audiologist who is comfortable with children. First, the audiologist must select a response behavior that the child is capable of performing. The second challenge is teaching the child to wait, listen, and only respond with the motor behavior when the auditory signal is presented. The third challenge is that the audiologist also must be skilled in delivering social reinforcement that is natural and rewarding at the appropriate time and interval. Separation of response behavior and reinforcement is essential in CPA. While the play activity is fun for the child, it is not the reinforcement. A separate reinforcement is essential to minimize habituation and maximize repeated response behavior. In addition to social praise, other forms of reinforcement have been suggested. Tokens that can be traded for small toys at the end of the test session, unsweetened cereal, and a changing computer display screen all have been used successfully with play audiometry. Audiologic literature suggests that CPA is widely accepted among clinicians who practice pediatric audiology (Thompson et al., 1989). It is generally recognized that most 3-year-olds can be tested using play audiometry. Yet, how young can children be to still achieve successful audiologic outcomes? Thompson and Weber (1974) demonstrated that the rate of success in obtaining detailed information with CPA is limited for children under the age of 30 months. However, some 2-year-olds can be conditioned to play audiometry (Thompson et al., 1989). In addition, when 2-year-olds are proficient with CPA, there is a greater likelihood that they will provide more responses before habituation than they would if tested by VRA. Because overlap exists between VRA and CPA as suitable techniques with children in this age range, the successful evaluation of a younger child with CPA ultimately depends on the following: the audiologist s observational skills of the child s developmental/maturational level, the interpersonal skills established between the audiologist and child, and the experience/comfort level of the audiologist with young children. Experience with CPA indicates that reliable threshold responses can be obtained when conditioning has been established and response criterion is maintained. Results from a clinical study (Diefendorf, 1981) of 40 preschoolers, aged 30 to 48 months, revealed thresholds at an audiometric level of 10 db HL or better. These findings were in close agreement with other 4-year-old children (Gerwin and Glorig, 1974). Striving to improve behavioral testing techniques is important because behavioral tests will continue to be the foundation of the audiologist s test battery. Moreover, behavioral tests provide the critical link between electrophysiologic measures and the child s daily use of audition. OTHER AUDIOMETRIC PROCEDURES FOR CHILDREN Because language and vocabulary are emerging in infants and young children, it may not be feasible to establish a traditional speech reception threshold (SRT). An alternative approach is the determination of a speech detection threshold (SDT). The SRT and SDT represent different criteria (intelligibility vs. detectability). The SRT is recognized as the intensity at which an individual is able to identify simple speech materials approximately 50% of the time. The SDT may be defined as the level at which a listener may just detect the presence of an ongoing speech signal (e.g., bai-bai-bai presented with an overall duration of approximately 2 seconds). Naturally, for a given individual, the threshold values will not be the same. Speech can be detected at intensity levels lower than it can be understood. This difference is on the order of 8 to 12 db. Once the audiologist has determined that the child is successful with a play audiometry task, obtaining an SRT for each ear may be useful (see Chapter 5 for a more detailed discussion of speech audiometry). This testing can be accomplished with headphones or insert earphones, if tolerated, or through the sound field speakers. The child who is ready for play audiometry typically has a communication strategy to express needs and wants at a more sophisticated level, whether with oral speech, signs, or a communication board. Family members often describe various communication skills that the child possesses, such as following commands, pointing to body parts or pictures in a storybook, or identifying colors. The audiologist is then able to expand the test battery to include an SRT rather than an SDT. A spondee picture board can be very helpful in obtaining an SRT from the child with special needs who may be reluctant to respond in an unfamiliar test situation. If the child uses a communication board, then items from the communication board can be selected to use for the SRT. Identification of body parts also can be used to obtain an SRT. Regardless of the test materials used, it is recommended that a preliminary step in determining an SRT for young children is first to familiarize the child with the test stimuli through both auditory and visual modalities and to eliminate those words that are not within the child s receptive vocabulary. The use of either picture or object pointing rather than a verbal response will require that the number of items be limited to 12 or less (Olsen and Matkin, 1979). Otherwise, the visual scanning task and the demands placed on memory and attention become contaminating variables. When obtaining an SRT, Northern and Downs (2002) suggest the use of a carrier phrase. They recommend a procedure in which the phrase show me is uttered and the hearing level is dropped quickly by 10 to 15 db for the test word. These authors suggest that the utilization of the carrier phrase, such as

16 Chapter 23 Assessment of Hearing Loss in Children 559 point to or show me, will often serve to focus the child s attention to the auditory task at hand. Finally, since a child s attention span is limited and test time can be a factor, it is often more expedient to work in 10-dB rather than 5-dB steps when establishing an SRT. The bone-conducted SRT can be extremely useful in obtaining additional data from children, and although it is typically underused, it is readily available to audiologists. Once a hearing loss is identified or suspected, any information that can be obtained regarding the type of hearing loss (conductive or sensory-neural) is helpful in the management of the child. Some children with special needs will be much less consistent for tonal stimuli, thereby compromising traditional bone-conducted puretone testing. However, the bone oscillator will deliver clear speech stimuli without any need for additional correction or modification. Dolan and Morris (1990) confirmed that the functions relating the percentage of correctly identified W-1 spondees to stimulus level and the slope of these functions were comparable for TDH-39 headphones and KH80 and B-72 bone vibrators. A bone-conducted SRT can offer valuable information in a very short period of time. Often the child will tolerate the bone oscillator during the more entertaining speech reception task but will not tolerate it for tonal testing. A frequently asked question regarding the use of the bone oscillator for speech reception testing relates to the potential for a false threshold that results in a vibratory response rather than a hearing response. It is true that the bone oscillator will vibrate for a speech stimulus, as well as low-frequency tonal stimuli, as the maximum output of the bone oscillator is approached. However, an important distinction must be made. A child will not be able to select the appropriate picture or item on the basis of a tactile sensation alone. If the child can complete the SRT, then a true hearing threshold by bone conduction has been obtained, and concerns regarding simply a vibratory response can be eliminated. The value of the bone-conducted SRT becomes even greater with the introduction of masking. Many youngsters become confused when masking is introduced during puretone testing. With the bone-conducted SRT, it is relatively easy to introduce masking into the nontest ear without interruption of the SRT procedure. Confirmation of a bilateral conductive component to a hearing loss is possible for many children who will not cooperate for masked puretone testing. Similarly, a unilateral sensory-neural or conductive hearing loss can be confirmed. The measurement of speech perception with the pediatric population must consider a number of variables that can confound test outcome. The selection of test materials within a child s receptive vocabulary competency, the designation of an appropriate response task, the utilization of reinforcement, and the reduction or alleviation of memory load are important factors that affect the reliability and validity of pediatric measurement. Haskins (1949) developed phonetically balance (PB) lists composed of monosyllabic words selected from the spoken vocabulary of kindergartners (PBK). The PBK lists of 50 words each have been widely used in working with children. Yet the receptive vocabulary level of the particular child under audiologic study is often not ascertained before administering these materials. Consequently, the PBK- 50 scores may be depressed in that they reflect vocabulary deficits as well as problems in speech perception. Clinicians must exercise caution in administering this test unless there is a relatively good assurance that the receptive vocabulary age of the child approaches at least that of a normal-hearing kindergartner. To bypass this problem, Ross and Lerman (1970) developed the Word Intelligibility by Picture Identification (WIPI) test. The WIPI test includes picture plates with six illustrations per plate. Four of the illustrations have words that rhyme, and the other two illustrations are presented as foils to decrease the probability of a correct guess. The use of WIPI materials is appropriate for those children with receptive vocabulary ages of 4 years and greater. There are differences between the PBK words and WIPI test approach to speech perception testing besides the evident fact that the latter is pictorially represented. PBK words represent an open response paradigm in which the child is forced to give a response from an unlimited set of possibilities, whereas the WIPI is a closed response set with the child s response being a forced choice (Kirk et al., 1997). As such, the use of the WIPI as a closed-set test improves the discrimination scores by about 10%. The Northwestern University-Children s Perception of Speech (NU-CHIPS) test by Elliott and Katz (1980) was developed as a speech perception test appropriate for younger children. Test materials are limited to monosyllabic words that are documented to be in the recognition vocabulary of children with normal hearing as young as age 3 years. Additionally, the authors report that children with hearing loss and a receptive language age of a least 2.6 years (as measured by the Peabody Picture Vocabulary Test; Dunn and Dunn, 1981) demonstrate familiarity with the words and pictures of the test. Continued surveillance for hearing impairment that may interfere with communication, development, health, or future academic performance must continue for preschool children. For this age group (3 to 5 years), screening for hearing impairment is a pass-refer procedure to identify individuals who require further audiologic evaluation or other assessments. Hearing impairment is defined as unilateral or bilateral sensory-neural and/or conductive hearing loss greater than 20 db HL in the frequency region from 1,000 to 4,000 Hz. Unlike the newborn and school-age populations, when nearly all children are accessible in hospitals and schools, preschoolers are generally not available in large, organized groups that lend themselves to universal screening for hearing impairment. As such, an interdisciplinary, collaborative effort is particularly important for this age group. Physicians and other professionals who make up the child s medical home and other professionals who specialize in child

17 560 Section III Special Populations development should be included in the planning and implementation of the hearing screening program to maximize the likelihood of prompt referral of children at risk of hearing impairment. SUMMARY Reaching the goal of early detection of hearing loss is facilitated by vigilant follow-up of infants and young children at risk for or suspected of having hearing loss. Furthermore, early detection of hearing loss is optimized by audiologists providing detailed audiologic assessment in a timely manner for those children referred from the screening process, early intervention lead agencies, the child s medical home, and other medical specialists. The accuracy and precision of our audiologic test battery also are critical in monitoring children with hearing loss and following children for delayed-onset hearing loss. Important and fundamental decisions in management and intervention depend on the audiometric outcomes and diagnosis provided by audiologists. If the development of this information is not accurate, precise, timely, and cost effective, we diminish the quality of services provided and compromise the credibility of our discipline. When these standards are met, audiologists positively impact the patients they serve and their families and are important members of the early multidisciplinary health care team. Additional text material for this chapter can be found at REFERENCES American Academy of Audiology. (2003) Pediatric amplification protocol. Available at: 53D26792-E321-41AF-850F-CC253310F9DB/0/pedamp.pdf. American Academy of Pediatrics. (2002) The medical home. Pediatrics. 110, American Speech-Language-Hearing Association. (2004) Guidelines for the audiologic assessment of children from birth to 5 years of age. Available at: deskref-journals/deskref/default Bayley N. (1969) Bayley Scales of Infant Development: Birth to Two Years. San Antonio, TX: Psychological Corp. Bess FH. (1982) Children with unilateral hearing loss. J Acad Rehabil Audiol. 15, Bess FH, Dodd-Murphy J, Parker RA. (1998) Children with minimal sensorineural hearing loss: prevalence, educational performance, and functional status. Ear Hear. 19, Bess FH, Klee T, Culbertson J. (1988) Identification, assessment and management of children with unilateral sensorineural hearing loss. Ear Hear. 7, Boege P, Janssen T. (2002) Pure-tone threshold estimation from extrapolated distortion product otoacoustic emission I/O functions in normal and cochlear hearing loss ears. J Acoust Soc Am. 111, Bovo R, Martini A, Agnoletto M, Beghi A, Carmignoto D, Milani M, Sangaglia AM. (1988) Auditory and academic performance of children with unilateral hearing loss. Scand Audiol. 30 (suppl), Carney A, Moeller MP. (1998) Treatment efficacy: hearing loss in children. J Speech Hear Res. 41, S61 S84. Condon MC. (1991) Unique challenges: children with multiple handicaps. In: Feigin J, Stelmachowicz P, eds. Pediatric Amplification. Omaha NE: Boys Town National Research Hospital. Culpepper B, Thompson G. (1994) Effects of reinforcer duration on the response behavior of preterm 2-year olds in visual reinforcement audiometry. Ear Hear. 15, Diefendorf AO. (1981). The effect of a pre-play period on play audiometry. Paper presented at Tennessee Speech and Hearing Association convention, Memphis, TN. Diefendorf AO. (1988) Behavioral evaluation of hearing-impaired children. In: Bess FH, ed. Children with Hearing Loss: Contemporary Trends. Nashville: Bill Wilkerson Center Press; pp Diefendorf AO, Gravel JS. (1996) Visual reinforcement and behavioral observation audiometry. In: Gerber SE, ed. Handbook of Pediatric Audiology. Washington, DC: Gallaudet University Press; pp Dolan TG, Morris SG. (1990) Administering audiometric speech tests via bone conduction: a comparison of transducers. Ear Hear. 11, Dorn PA, Konrad-Martin D, Neely ST, Keefe DH, Cry E, Gorga MP. (2001) Distortion product otoacoustic emission input/output functions in normal-hearing and hearing-impaired human ears. J Acoust Soc Am. 110, Dunn L, Dunn L. (1981) Peabody Picture Vocabulary Test Revised. Circle Pines, MN: American Guidance Service. Eilers RE, Miskiel E, Ozdamar O, Urbano R, Widen J. (1991a) Optimization of automated hearing test algorithms: simulations using an infant response model. Ear Hear. 12, Eilers RE, Widen J, Urbano R, Hudson TM, Gonzales L. (1991b) Optimization of automated hearing test algorithms: a comparison of data from simulations and young children. Ear Hear. 12, Elliott LL, Katz D. (1980) Development of a New Children s Test of Speech Discrimination (Technical Manual). St Louis, MO: Auditec. Finitzo T, Crumley WG. (1999) The role of the pediatrician in hearing loss from detection to connection. Pediatr Clin North Am. 46, Genetic Evaluation of Congenital Hearing Loss Expert Panel. (2002) Genetics evaluation guidelines for the etiologic diagnosis of congenital hearing loss. Genet Med. 4, Gerwin KS, Glorig A, eds. (1974) Detection of Hearing Loss and Ear Disease in Children. Springfield, IL: Charles C. Thomas. Goldstein K, Caputo D, Taub H. (1976) The affects of prenatal and perinatal complications on development at one year of age. Child Dev. 47,

18 Chapter 23 Assessment of Hearing Loss in Children 561 Gorga MP, Neely ST, Bergman B, Beauchaine K, Kaminski J, Peters J, Jesteadt W. (1993) Otoacoustic emissions from normalhearing and hearing-impaired subjects: distortion product responses. J Acoust Soc Am. 93, Gorga MP, Neely ST, Dierking DM, Dorn PA, Hoover BM, Fitzpatrick D. (2003) Distortion product otoacoustic emission tuning curves in normal-hearing and hearing-impaired human ears. J Acoust Soc Am. 114, Gorga MP, Neely ST, Dorn PA. (2002) Distortion product otoacoustic emissions in relation to hearing loss. In: Robinette MS, Glattke TJ, eds. Otoacoustic Emissions: Clinical Applications. 2nd ed. New York: Thieme Medical; pp Gorga MP, Stover LT, Neely ST. (1996) The use of cumulative distributions to determine critical values and levels of confidence for clinical distortion product otoacoustic emission measurements. J Acoust Soc Am. 100, Gorlin RJ, Toriello HV, Cohen MM. (1995) Hereditary Hearing Loss and Its Syndromes. New York: Oxford University Press. Gravel JS, Traquina DN. (1992) Experience with the audiological assessment of infants and toddlers. Int J Pediatr Otolaryngol. 23, Gravel JS, Wallace IF. (1999) Otitis media and communication during pre-school years. Paper presented at ICIS2000 Doctorial Consortium, Brisbane, Australia. Greenburg D, Wilson WR, Moore JM, Thompson G. (1978) Visual reinforcement audiometry (VRA) with young Down syndrome children. J Speech Hear Dis. 43, Hanley M. (1986) Basic Principles of Auditory Assessment. San Diego: College Hill Press. Haskins H. (1949) A phonetically balanced test of speech discrimination for children. Master s thesis, Northwestern University, Evanston, IL. Jerger J, Hayes D. (1976) The cross-check principle in pediatric audiometry. Arch Otolaryngol. 102, John MS, Brown DK, Muir PJ, Picton TW. (2004) Recording auditory steady-state responses in young infants. Ear Hear. 25, Joint Committee on Infant Hearing (JCIH). (1972) Supplementary statement on infant hearing screening. ASHA. 16, 160. Joint Committee on Infant Hearing (JCIH). (1982) 1982 position statement. ASHA. 24, Joint Committee on Infant Hearing (JCIH). (1991) 1990 position statement. ASHA. 33, 3 6. Joint Committee on Infant Hearing (JCIH). (1994) Position statement. ASHA. 36, Joint Committee on Infant Hearing (JCIH). (2000) Year 2000 position statement: principles and guidelines for early hearing detection and intervention programs. Am J Audiol. 9, Joint Committee on Infant Hearing (JCIH). (2007) Year 2007 position statement: principles and guidelines for early hearing detection and intervention programs. Pediatrics. 120, Katbamna B, Homnick DN, Marks JH. (1999) Effects of chronic tobramycin treatment on distortion product otoacoustic emissions. Ear Hear. 20, Kemp DT, Ryan S, Bray P. (1990) A guide to the effective use of otoacoustic emissions. Ear Hear. 11, Kirk KI, Diefendorf AO, Pisoni DB, Robbins AM. (1997) Assessing Speech Perception in Children. In: Mendel LL, Danhauer JL, eds. Audiologic Evaluation and Management and Speech Perception Assessment. San Diego: Singular Publishing Group, Kopp C. (1974) Fine motor abilities of infants. Dev Med Child Neurol. 16, Lonsbury-Martin BL, McCoy MJ, Whitehead ML, Martin GK. (1993) Clinical testing of distortion-product otoacoustic emissions. Ear Hear. 14, Margolis RH, Bass-Ringdahl S, Hanks WD, Holte K, Zapala DA. (2003) Tympanometry in newborn infants 1 khz norms. J Am Acad Audiol. 14, Martin GK, Ohlms LA, Franklin DJ, Harris FP, Lonsbury-Martin BL. (1990) Distortion product emissions in humans III. Influence of sensorineural hearing loss. Ann Otol Rhinol Laryngol Suppl. 147, Mehl AL, Thomson V. (1998) Newborn hearing screening: the great omission. Pediatrics E4. Meredith R, Stephens D, Sirimanna T, Meyer-Bisch C, Reardon W. (1992) Audiometric detection of carrier of Usher s syndrome type II. J Audiol Med. 1, Moeller MP. (2000) Early intervention and language outcomes in children who are deaf and hard of hearing. Pediatrics. 106, 1 9. Moore JM. (1995) Behavioral assessment procedures based on conditioned head-turn responses for auditory detection and discrimination with low-functioning children. Scan Audiol Suppl. 41, Moore JM, Thompson G, Folsom R. (1992). Auditory responsiveness of premature infants utilizing visual reinforcement audiometry (VRA). Ear Hear. 13, Moore JM, Thompson G, Thompson M. (1975) Auditory localization of infants as a function of reinforcement conditions. J Speech Hear Disord. 40, Moore JM, Wilson WR, Thompson G. (1977) Visual reinforcement of head-turn responses in infants under 12 months of age. J Speech Hear Disord. 40, Morton C. (1999) The NIDCD Working Group on genetic testing for deafness and other communication disorders: considerations for developing and implementing testing. Presented at Workshop on the Genetics of Congenital Hearing Impairment, Centers for Disease Control and Prevention and Gallaudet University, June 7, 1999, Atlanta, GA. Nance WE. (2003) The genetics of deafness. Ment Retard Dev Disabil Res Rev. 9, National Center for Hearing Assessment and Management. (2007) EDHI legislation. Available at: legislative/index.html. Northern JL, Downs MP. (2002) Hearing in Children. Baltimore: Lippincott Williams & Wilkins. Norton SJ, Widen JE. (1990) Evoked otoacoustic emissions in normal-hearing infants and children: emerging data and issues. Ear Hear. 11, Nozza R, Wilson WR. (1984) Masked and unmasked pure tone thresholds of infants and adults: development of auditory frequency selectivity and sensitivity. J Speech Hear Res. 27, Olsen WO, Matkin ND. (1979) Speech audiometry. In: Rintelmann WF, ed. Hearing Assessment. Baltimore: University Park Press. Oyler R, Oyler A, Matkin ND. (1988) Demographics and educational impact. Lang Speech Hear Services Schools. 19, Paradise J, Smith DG, Bluestone CD. (1976) Tympanometric detection of middle ear effusion in infants and young children. Pediatrics. 58,

19 562 Section III Special Populations Primus M. (1992) Operant response in infants as a function of time interval following signal onset. J Speech Hear Res. 35, Primus M, Thompson G. (1985) Response strength of young children in operant audiometry. J Speech Hear Res. 18, Purdy SC, Williams MJ. (2000) High frequency tympanometry: a valid and reliable immittance test protocol for young infants? N Z Audiol Soc Bull. 10, Renshaw JJ, Diefendorf AO. (1998) Adapting the test battery for the child with special needs. In: Bess FH, ed. Children with Hearing Impairment. Nashville, TN: Vanderbilt Bill Wilkerson Press; pp Ress BD, Sridhar KS, Balkany TJ, Waxman GM, Stagner BB, Lonsbury-Martin BL. (1999) Effects of cis-platinum chemotherapy on otoacoustic emissions. The development of an objective screening protocol. Otolaryngol Head Neck Surg. 121, Ross M, Lerman J. (1979) Picture identification test for hearingimpaired children. J Speech Hear Res. 13, Rubin R, Rosenblatt C, Balow B. (1973) Psychological and educational sequelae of prematurity. Pediatrics. 52, Schmida MJ, Peterson HJ, Tharpe AM. (2003) Visual reinforcement audiometry using digital video disc and conventional reinforcers. Am J Audiol. 12, Seewald RC, Cornelisse LE, Ramji KV, Sinclair ST, Moodie SK, Jamieson DG. (1997) DSL v4.1 for Windows: A Software Implementation of the Desired Sensation Level (DSL[i/o]) Method for Fitting Linear Gain and Wide-Dynamic-Range Compression Hearing Instruments. User s manual. London, Ontario, Canada: University of Western Ontario. Stapells DR, Gravel JS, Martin BA. (1995) Thresholds for auditory brainstem responses to tones in notched noise from infants and young children with normal hearing or sensorineural hearing loss. Ear Hear. 16, Stapells DR, Oates P. (1997) Estimation of the pure-tone audiogram by the auditory brainstem response: a review. Audiol Neurootol. 2, Starr A, Picton TW, Sininger Y, Hood LJ, Berlin CI. (1996) Auditory neuropathy. Brain. 119, Stelmachowicz PG. (2000) How do we know we ve got it right? Electroacoustic and audiometric measures. In: Seewald R, ed. A Sound Foundation through Early Amplification: Proceedings of an International Conference. Stafa, Switzerland: PHONAK AG. The Pediatric Working Group. (1996) Amplification for Infants and Children with Hearing Loss. Nashville, TN: Vanderbilt Bill Wilkerson Press. Thompson G, Folsom RC. (1984) A comparison of two conditioning procedures in the use of visual reinforcement audiometry (VRA). J Speech Hear Disord. 49, Thompson G, Weber B. (1974) Responses of infants and young children to behavioral observation audiometry (BOA). J Speech Hear Disord. 39, Thompson G, Wilson WR, Moore JM. (1979) Application of visual reinforcement audiometry (VRA) to low-functioning children. J Speech Hear Disord. 54, Thompson MD, Thompson G, Vethivelu S. (1989) A comparison of audiometric test thresholds for 2-year-old children. J Speech Hear Disord. 54, Tonniges TF, Palfrey JS. (2004) The medical home. Pediatrics. 113, Turner RG. (2003) Double checking the cross-check principle. JAm Acad Audiol. 14, US Department of Health and Human Services. (2000) Healthy People nd ed. Washington, DC: US Government Printing Office. Weber BA. (1969) Validation of observer judgments in behavior observation audiometry. J Speech Hear Disord. 34, Widen JD. (1990) Behavioral screening of high-risk infants using visual reinforcement audiometry. Semin Hear. 11, Wilson WR, Folsom RC, Widen JE. (1983) Hearing impairment in Down s syndrome children. In: Mencher G, Gerber S, eds. The Multiply Handicapped Hearing Impaired Child.NewYork: Grune & Stratton. Wilson WR, Moore J, Thompson G. (1976) Sound-field auditory thresholds of infants utilizing visual reinforcement audiometry (VRA). Paper presented at the ASHA Annual Convention, November 20 23, 1976, Houston, TX. Yang EY, Stewart A. (1990) A method of auditory brain stem response to bone conducted clicks in testing infants. J Speech Lang Pathol Audiol. 14, Yoshinaga-Itano C. (1995) Efficacy of early identification and intervention. Semin Hear. 16, Yoshinaga-Itano C. (2003). From screening to early identification and intervention: discovering predictors to successful outcomes for children with significant hearing loss. J Deaf Stud Deaf Educ. 8, Yoshinaga-Itano C, Sedey A. (2000) Development of audition and speech: implications for early intervention with infants who are deaf or hard of hearing. In: Yoshinaga-Itano C, Sedey AL, eds. Language, Speech and Social-Emotional Development of Children Who Are Deaf and Hard of Hearing: The Early Years. Volta Rev. 100, Yoshinaga-Itano C, Sedey A, Coulter DK, Mehl AL. (1998) Language of early and later identified children with hearing loss. Pediatrics. 102,

20 APPENDIX 23.1 Risk Indicators Associated with Permanent Early-Onset and/or Late Progressive Hearing Loss in Children (Joint Committee on Infant Hearing, 2007) 1. Caregiver concern regarding hearing, speech, language, or developmental delay 2. Family history of permanent childhood hearing loss 3. Neonatal intensive care of >5 days, assisted ventilation 10 days, prolonged exposure to ototoxic medications 7 days (gentamicin and tobramycin) or loop diuretics (furosemide/lasix), hyperbilirubinemia requiring exchange transfusion and extracorporeal membrane oxygenation (ECMO) 4. In utero infections such as cytomegalovirus, herpes, rubella, syphilis, and toxoplasmosis 5. Craniofacial anomalies, including those involving the pinna and ear canal, ear tags, ear pits, and temporal bone anomalies 6. Physical findings such as white forelock, associated with a syndrome known to include a sensory-neural or permanent conductive hearing loss 7. Syndromes associated with progressive hearing loss such as neurofibromatosis, osteopetrosis, and Usher syndrome 8. Neurodegenerative disorders, such as Hunter syndrome, or sensory motor neuropathies, such as Friedreich s ataxia and Charcot-Marie-Tooth syndrome 9. Culture-positive postnatal infections associated with sensory-neural hearing loss, including confirmed bacterial and viral (especially herpes viruses and varicella) meningitis 10. Head trauma, especially basal skull/temporal bone fracture, requiring hospitalization 11. Chemotherapy Risk indicators that are of greater concern for delayed-onset hearing loss. 563