Objectives. There is a paucity of data describing the epidemiology of serious hearing impairment among children in the United States. This report provides information on the prevalence of serious hearing impairment among children born in the 1980s and living in the metropolitan Atlanta area in 1991–1993 and on the characteristics of children with serious hearing impairment.
Methods. Data for this report are drawn from the Metropolitan Atlanta Developmental Disabilities Surveillance Program, an ongoing, active case-ascertainment system for mental retardation, cerebral palsy, hearing impairment, and vision impairment among children 3 to 10 years of age. Hearing impairment was defined as a bilateral, pure-tone hearing loss at frequencies of 500, 1000, and 2000 Hz averaging 40 decibels or more, unaided, in the better ear. Both severity and type of hearing loss were examined. Cross-sectional as well as birth cohort prevalence rates of serious hearing impairment were computed by sex and by race. The presence of mental retardation, cerebral palsy, vision impairment, or a seizure disorder was also assessed. An attempt was made to determine the probable etiology of a subset of the cases.
Results. The average, annual prevalence rate for moderate to profound hearing loss was 1.1 per 1000. The prevalence rate increased steadily with age. Ninety percent of all cases for which the type of loss was recorded were sensorineural. The highest rate was seen among black male children (1.4 per 1000). Thirty percent of case children had another neurodevelopmental condition, most frequently mental retardation. Black male children also experienced the highest rate of presumed congenital hearing impairment. The mean age at which children with presumed congenital hearing impairment first met the surveillance case definition was 2.9 years. A probable etiology could only be found for 22% of cases born in the study area.
Conclusions. The data presented here provide information on the descriptive epidemiology of serious hearing impairment among United States children. The reasons for the higher rates among black children, especially males, may be a fruitful direction for further research.
Hearing loss among children is a relatively common and serious problem that can have a significant impact on their development. The extent and severity of loss can range from unilateral and mild (26–40 decibel [dB] loss in the affected ear) to bilateral and profound (>85 dB loss in the better ear).1 Mild hearing loss may warrant retesting, placement of a child's desk in the front of the classroom, speech therapy, or wearing a hearing aid in school, and speech and language skills are usually normal or slightly impaired.2 A child with moderate hearing loss (∼40–64 dB) can understand speech 3 to 5 feet away and requires hearing amplification, speech and reading training, and therapy.1A child with severe hearing loss (∼65–84 dB) can understand loud speech no more than a foot away and requires special education services. A child with profound hearing loss does not rely on hearing as his or her primary modality for communication.1
The age of onset of a child's hearing impairment and the age at which the hearing impairment is diagnosed are crucial to the child's speech-language, cognitive, and psychosocial development.3,,4 The older a child is when a congenital hearing loss is diagnosed, the greater the child's chance of developmental delay.5 Previous protocols for screening newborns for hearing loss based on certain high-risk criteria usually identify only ∼50% of all infants with congenital hearing loss.3,,6,7 The remainder may not be identified early enough, often resulting in interventions being delayed past the crucial time of communication development.
In previous epidemiologic studies of the prevalence of hearing loss, investigators have used varying definitions of hearing impairment, making it difficult to compare the results. The variation in prevalence rates found in previous studies8 (from ∼1 to 3 cases per 1000 children) is primarily the result of differences in the age ranges of the children studied, the extent of loss (unilateral or bilateral) required to meet the case definition, the severity levels included (slight, mild, mild/moderate, moderate, severe, profound) and their accompanying decibel definitions, the method used to identify cases, and whether the rates presented are based on the number of live births or the number of children in the population in a given age range.
In this article, we use data from 1991–1993 from an ongoing, prospective surveillance system, the Metropolitan Atlanta Developmental Disabilities Surveillance Program (MADDSP), to examine four features of serious hearing impairment: its overall prevalence, its descriptive epidemiology (sex, race, type, and level of hearing loss), its association with other developmental disabilities, probable etiology, and the distribution of children's ages at the time of the earliest known diagnosis of congenital hearing impairment.
MATERIALS AND METHODS
Established in 1991, MADDSP is an ongoing system for monitoring the prevalence of selected developmental disabilities: mental retardation, cerebral palsy, serious hearing impairment, and vision impairment.9 MADDSP seeks to ascertain all affected children ages 3 to 10 years whose parent(s) or legal guardian(s) are residents of one of five Georgia counties (Clayton, Cobb, Dekalb, Fulton, and Gwinnett) during the study year of interest. Existing records are reviewed at multiple sources to identify children who meet the surveillance case definition for one or more of the conditions of interest. The surveillance case definition for serious hearing impairment is a bilateral, pure-tone hearing loss at frequencies of 500, 1000, and 2000 Hz averaging 40 dB or more, unaided, in the better ear. Case definitions for the other three conditions have been published elsewhere.9 We chose 40 dB as the threshold value for MADDSP because we were reasonably certain that the vast majority of such children would be receiving services in the public school systems and thus be easily identifiable.
Children with serious hearing impairment were ascertained at facilities providing services for children with special needs, including special education programs at the nine public school systems serving the study area and other programs sponsored by the Georgia Department of Education, specifically state schools for children with hearing impairment. Additional sources for identifying eligible children include the Georgia Department of Human Resources facilities that provide services to children with developmental disabilities, as well as one public and two private pediatric hospitals and their associated clinics in the metropolitan Atlanta area. Ninety-five percent of children with hearing impairment were identified from public and state schools, 3% from public and private hospitals and their associated clinics, and 2% from the Department of Human Resources.
Data collected on all seriously hearing-impaired children included the type and level of hearing loss and the presence of other medical conditions, such as seizure disorder. The type of hearing loss was categorized as sensorineural, conductive, mixed, or unknown. Sensorineural hearing loss is attributable to an abnormal condition of the inner ear or neural pathology of the central nervous system that cannot be resolved completely by medical intervention. Conductive hearing loss is the result of reduced conduction of sound through the outer and/or middle ear to the inner ear. Although, most cases of conductive hearing loss can be corrected by medical or surgical treatment, some conductive pathologies may persist and require amplification and/or educational intervention. Mixed hearing loss involves elements of both conductive and sensorineural hearing loss and may be treated medically and/or with hearing amplification and educational intervention. The levels of hearing impairment defined in MADDSP are moderate (40–64 dB), severe (65–84 dB), and profound (≥85 dB). For this report, we categorized case children by whether they also had mental retardation, cerebral palsy, vision impairment, or a seizure disorder. This categorization allowed us to establish a subgroup of children with hearing impairment but no other serious neurodevelopmental conditions, referred to here as isolated hearing impairment.
The methodology and results of this article are separated into two sections: 1) cross-sectional analyses, in which we counted all children with serious hearing impairment to compute average, annual prevalence rates; and 2) birth cohort analyses, in which we divided the number of children with serious hearing impairment born to residents of the study area by the number of live births in the study area to compute birth prevalence rates.
Postcensal estimates provided by the Bureau of the Census for Georgia indicated that 255 742 children ages 3 to 10 years lived in the metropolitan Atlanta area during 1991, 262 745 in 1992, and 271 713 in 1993. During this period, MADDSP identified 283 children who met the case definition for serious hearing impairment in 1991, 288 in 1992, and 291 in 1993—in all, 411 different children. For this report, we defined race as either white or black because 398 (97%) of the 411 case children were either black or white. The category of white children, in accordance with Bureau of the Census practice, includes both white and Hispanic children.
We calculated average, annual prevalence rates of hearing impairment for 1991–1993 by dividing the sum of the annual numbers of case children (n = 862) by the sum of the annual census populations for children ages 3 to 10 years living in the metropolitan Atlanta area (790 200). We computed 95% confidence intervals for the prevalence rates by using the exact binomial method.10 All case children were used in most analyses, except for those analyses involving type of hearing loss, because the type of loss was unknown for 57 (13.9%) of the 411 case children.
Birth Cohort Analyses
We restricted the case population for computing birth prevalence rates by sex and race to the children whose mothers resided within the metropolitan Atlanta ascertainment area at the time of their child's birth (n = 250). A panel of three physicians independently reviewed all available medical data and identified a subset of these case children (n = 227) whose hearing impairment was not clearly the result of a postneonatal event. The data reviewed included data from the Metropolitan Atlanta Congenital Defects Program, an active birth defects surveillance system covering the same ascertainment area as MADDSP.11 The medical review panel also attempted to assign a probable etiology for each of the 250 case children. To identify the group of case children with the highest probability of having congenital hearing impairment, we limited our study population to sensorineural cases with no stated postneonatal event as the probable etiology (n = 173). We considered these children to have a presumed congenital hearing impairment. The earliest age at diagnosis for these 173 children was defined as the earliest known age at which records available to MADDSP indicated that the child met the surveillance case definition for hearing impairment.
Data from Georgia vital statistics indicated that 324 327 children were born alive between 1981 and 1990 to residents of the study area. This figure served as the denominator for the overall birth prevalence rate.
The overall prevalence rate of serious hearing impairment was 1.10 cases per 1000 children ages 3 to 10 years during the period 1991–1993. The overall prevalence rate was the same for each of the 3 individual study years. The prevalence rate was lowest among 3-year-old children (0.67) and increased steadily to a high of 1.38 among 10-year-old children (Table 1). The distribution of case children by type and severity of loss is shown inTable 2. Almost 90% of all case children with a known type of loss had sensorineural loss. The prevalence rates of moderate, severe, and profound loss were 0.48, 0.25, and 0.37 per 1000, respectively.
The prevalence rate of hearing impairment among males (1.23 per 1000) was higher than that among females (0.95 per 1000), and the prevalence rate among black children was higher than that among white children (Table 3). The highest rate was that for black males (1.40 per 1000). When we limited the analysis to those children with sensorineural hearing loss, black males still had the highest prevalence rate of the four sex-race subgroups (Table 3). Further, black males exhibited the highest prevalence rate at each of the three levels of severity.
More than two-thirds (69.6%) of the 411 children with hearing impairment had no other known developmental disability, whereas 30.4% had at least one other developmental disability (Table 4). Mental retardation was the most common of the other four developmental disabilities (26.3%). The proportion of hearing-impaired children with isolated hearing impairment did not vary markedly by severity of hearing loss (P = .26). Although there was no significant difference in the proportion of children with isolated hearing impairment among the three known types of hearing loss (sensorineural = 75.7%, conductive = 68.0%, and mixed = 83.3%), only 33.3% of the 57 case children with an unknown type of hearing loss had no other developmental disability.
Birth Cohort Analyses
Of the 250 case children, a probable etiology was found in only 22% (Table 5).
The overall birth prevalence rate of presumed congenital hearing impairment was 0.53 cases per 1000 live births. The distribution of severity levels for the 173 children with this form of hearing impairment was moderate, 39.3%; severe, 24.9%; and profound, 35.8%. Black males had the highest birth prevalence rate (Table 6), as well as the highest birth prevalence rate at each severity level. In contrast, there were no appreciable differences in birth prevalence rates of serious hearing impairment presumed to have been postnatally acquired among the four sex-race subgroups (Table 6).
Among the 173 case children with presumed congenital hearing impairment, the mean age at earliest known diagnosis was 2.9 years. Of these children, 13 (7.5%) had their hearing impairment diagnosed within the first year of life, and 80 (46.5%) did not have their hearing impairment diagnosed until at least 3 years of age (Fig 1). The mean age at diagnosis for children with moderate hearing loss was 3.6 years compared with 2.4 years each for children with severe or profound hearing loss. Severity levels did not differ significantly between black and white children nor did mean age at earliest known diagnosis. The mean age at diagnosis for those children with isolated hearing impairment (n= 121) was 2.6 years compared with 3.4 years of age for those case children with multiple developmental disabilities (n = 52). This difference in mean age at diagnosis was not attributable to differences in severity levels (P = .45).
Several findings from this study provide important information on the occurrence of serious hearing impairment among American children. First, we documented a higher-than-average rate of hearing impairment among black male children that prevails across all severity levels and that is presumably congenital in origin. Second, we found that almost half of the children born in the Atlanta metropolitan area in the 1980s with serious hearing impairment present at birth were not being recognized as seriously hearing impaired until they were at least 3 years old.
As a point of reference, the present results can be placed in the context of the more complete distribution of hearing loss obtained from audiometric screening of a nationwide sample of children 6 to 19 years old done as a part of the Third National Health and Nutrition Examination Survey (NHANES III).12 Those data indicated that an estimated 4 children per 1000 had at least mild (>25 dB), bilateral, low-frequency loss and that 3 per 1000 had low-frequency loss in either ear that was severe to profound (>65 dB). In light of the differences in the definitions of hearing impairment between NHANES III and MADDSP, it is worth noting that our highest prevalence rate (ie, among 10-year-old children) was still less than half of either of the aforementioned rates. Because of the nature of NHANES III, it is difficult to determine which participants had temporary versus permanent hearing loss or to determine if children meeting the MADDSP case definition would have been differentially included or excluded in the national survey.
The higher rates of hearing impairment among black children were consistent with rates observed in an earlier study of 10-year-old children in the Atlanta area13 and could be partly because of racial differences in the prevalence of low birth weight and to differential exposure to known etiologic agents. Subsequent analyses of the MADDSP data will examine racial differences in birth weight-specific prevalence rates of serious hearing loss and in probable etiologies.
A previous study of hearing impairment in the Atlanta area, involving children who were born from 1975 to 1977 and using virtually the same case-finding methods and case definition as ours, found a point prevalence rate of 1.1 per 1000 10-year-old children,13compared with the rate of 1.38 per 1000 that we found. Possible reasons for this difference, besides the possibility that it is attributable to chance, include an increase in the proportion of black residents in the metropolitan Atlanta area, an increase in the proportion of low birth weight infants, an increase in the prevalence of one or more etiologic agents, and the interruption of intervention strategies for the known causes of hearing impairment. The first two possible reasons seem unlikely in view of the short time span (6 years) between the two studies.
Although the prevalence of multiple developmental disabilities among the case children did not differ by the type or severity of their hearing loss, we found a much higher prevalence of multiple developmental disabilities among the 57 children with an unknown type of hearing loss. Approximately half of these 57 case children had a low birth weight (<2500 g) and approximately half had cerebral palsy, the latter being a significantly higher proportion than among all case children (P < .005). Further, these 57 children, as a group, had milder levels of hearing loss. The reasons for the unusual composition of this subgroup of case children are unclear.
Based on the information available to us, we could assign a probable etiology to only 22% of 250 case children. The large proportion of idiopathic cases partly reflects the fact that MADDSP is limited to review of existing records and does not have access to either hospital newborn records or to records of private providers for hearing impaired children. As newborn hearing screening programs become more widespread, there may be opportunities to gather more etiologic data in the future.
Of the 173 study children with presumed congenital hearing impairment, the available data indicated that nearly half were at least 3 years old before their condition was diagnosed. A word of caution may be in order, however, regarding this finding. Because record review in MADDSP is done retrospectively, we could have missed an earlier record of hearing loss for some children. Thus, for some children, the age at earliest diagnosis in MADDSP may not be the actual earliest age at which some degree of hearing loss was detected. This may partly explain why the mean age at earliest known diagnosis seen here is much higher than that reported in a Danish case series (∼1.4 years) of similar age and birth year range as our cohort, although the Danish study included milder degrees of hearing loss.14 We could not examine changes in mean age at diagnosis throughout time (ie, birth cohorts), because age at ascertainment varied inversely with birth year.
We found that approximately one-third of the hearing-impaired children identified in MADDSP had one or more other developmental disabilities. Our finding that 26% of children with hearing impairment also had mental retardation is higher than that found by Elliott15(17%), but it is similar to the proportion found in a previous study in the Atlanta area (21%).13 These results suggest that medical and other professionals who interact with seriously hearing-impaired children may need to be aware of their cognitive status.
In calculating birth prevalence rates, we chose to use the number of live births as the denominator rather than the number of children who had not died before age 3 years, the minimum age for inclusion as a case child in MADDSP. We selected this method primarily so that our rates would be comparable with those of other studies. In any case, our choice of denominator had little effect on the resulting rate. The difference between our overall point prevalence rate (1.10 per 1000) and our overall birth prevalence rate (0.53 per 1000) is most likely because of out-migration of children from the birth cohorts.
Our birth prevalence rate of presumed congenital hearing impairment among white children in metropolitan Atlanta (0.41 per 1000) is somewhat lower than comparable rates obtained in Göteborg, Sweden16 and in one eastern province in Finland17 and is dramatically lower than the birth prevalence rate of 1.2 per 1000 found in a joint Denmark-England study.18 Whether marked discrepancy between our results and those of the latter study is real or because of differences in types of hearing loss included, in emigration rates, or in ascertainment methods is unclear.
The strengths of this study include the size of the case series, the use of objective test data derived from multiple sources, and an active case ascertainment procedure. For these reasons, we believe that our prevalence rates are more indicative of the actual descriptive epidemiology of serious hearing impairment in the study area than rates derived from surveys dependent on passive reporting, on the use of only one or two sources, or on parental self-report. Further, because the surveillance area comprises a large urban center with racial diversity, the findings may be relevant to many other areas in the country. Conversely, there are some limitations in the way MADDSP is conducted. For example, because the surveillance data are obtained only from existing records to which the surveillance staff has access, some information relevant to a child's disability (eg, earliest age of diagnosis, etiology) may not be found.
MADDSP is currently the only ongoing source of information on the prevalence of serious hearing loss among children in the United States that is not derived from a one-time hearing screening test. The results presented here indicate that examination of the sources of racial differences in the occurrences of serious hearing impairment may be a fruitful direction for further research.
We thank Dr Marshalyn Yeargin-Allsopp and Dr Sarah Winter of the Centers for Disease Control and Prevention, Division of Birth Defects and Developmental Disabilities, for their contributions on our physician review panel.
- Received June 15, 1998.
- Accepted August 24, 1998.
Reprint requests to (P.D.) Centers for Disease Control and Prevention (Mailstop F-15), 4770 Buford Hwy NE, Atlanta, GA 30341.
- dB =
- decibel •
- MADDSP =
- Metropolitan Atlanta Developmental Disabilities Surveillance Program •
- NHANES III =
- Third National Health and Nutrition Examination Survey
- ↵Roeser RJ. Audiometric and immittance measures: principles and interpretation. In: Roeser RJ, ed. Auditory Disorders in School Children. 2nd ed. New York, NY: Thieme Medical Publishers; 1988:1–34
- ↵Rapin I. Children with hearing impairment. In: Swaiman KF, ed. Pediatric Neurology, II. St Louis, MO: CV Mosby Company; 1989:895–908
- Elssmann SF,
- Matkin ND,
- Sabo MP
- Pappas DG
- ↵Centers for Disease Control and Prevention. Prevalence of selected developmental disabilities in children 3–10 years of age: the Metropolitan Atlanta Developmental Disabilities Surveillance Program, 1991. MMWR. 1996;45(SS-2):1–13
- ↵Fleiss JL. Statistical Methods for Rates and Proportions. 2nd ed. New York, NY: John Wiley and Sons; 1981:14–15
- Edmonds LD,
- Layde PM,
- James LM,
- Flynt JE,
- Erickson JD,
- Oakley GP Jr.
- Davis A,
- Parving A
- Copyright © 1999 American Academy of Pediatrics