| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Departments of 1 Epidemiology and Public Health
2 Pediatrics
3 Ophthalmology, University of Miami School of Medicine, Florida.
Address correspondence to David J. Lee, PhD, Department of Epidemiology & Public Health, University of Miami School of Medicine, P.O. Box 016069 (R-699), Miami, FL 33101. E-mail: dlee{at}med.miami.edu
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. The National Health Interview Survey is a continuous multistage area probability survey of the U.S. civilian noninstitutionalized population living at addressed dwellings. Adults within randomly selected households were administered a chronic conditions list that included questions about hearing impairment. Complete data were available on 107,100 white and 17,904 African-American adults aged 18 years and older in survey years 1986–1995. Race-specific rates of hearing impairment were adjusted for age and sample survey design.
Results. Annual age-adjusted rates of some hearing impairment ranged from 11.0% to 12.7% in whites and 5.9% to 8.5% in African Americans. Rates of severe bilateral hearing impairment in these race groups were 0.7% to 1.1% and 0.1% to 0.5%, respectively. There was no evidence of change in rates of hearing impairment among participants stratified by race and 10-year age groups.
Conclusions. Reported rates of hearing impairment remained relatively stable in the U.S. noninstitutionalized population from 1986 to 1995. There was no evidence of change in rates in adults grouped into 10-year age groups. Population-based studies designed to include clinical and self-reported measures of hearing impairment are needed to further examine trends in hearing impairment.
There has also been concern, but thus far no conclusive evidence, that increases in recreational noise exposure may lead to rising rates of hearing impairment in the adult population. For example, exposure to amplified music is a relatively recent cultural phenomenon (7) and evidence of noise-induced hearing loss in children and adolescents residing in the United States has recently been reported (8). However, the cumulative effects of recreational noise exposure may not be apparent to individuals for several decades later, as further age-associated declines in hearing acuity occur and noise-induced damage extends further into frequencies essential for speech comprehension (9). Thus, any increase in reported hearing impairment rates due to a rise in environmental noise exposure levels may not be evident until middle age.
Clinical data to determine if hearing impairment rates are changing are not available since there are no ongoing, clinical, annual studies of hearing impairment in the United States. However, the National Center for Health Statistics (NCHS) has included questions about hearing impairment in its annual National Health Interview Survey (NHIS) in the years 1986–1995. Survey modifications after 1995 limited data comparability with earlier survey years. Thus, NHIS survey years 1986–1995 represent the best, most recently available data on 10-year trends in reported hearing impairment in the United States.
|
METHODS |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Assessment of Hearing Impairment
A random one in six subsample of households was administered a chronic condition list that included two hearing impairment questions. Participants were asked to indicate if they or any of their family members were deaf in one or both ears or had any other trouble hearing with one or both ears. In most cases, the participants themselves answered all the questions (61%–63% across survey years), and for the remaining participants, the responses were obtained from their relatives or other proxies. For simplicity, in the present study, self-reported or proxy-reported data are referred to as "reported." A series of standardized questions was used to detail the characteristics, cause, onset, and effects of each reported hearing impairment (11). Trained medical coders used this information and the NCHS-modified International Classification of Diseases, Ninth Revision (ICD-9) to classify impairment as: 1) deaf in both ears, 2) hearing impairment in one or both ears, 3) deaf in one ear and hearing impairment in the other ear, and 4) deaf or hearing impaired in one ear only (other ear, good hearing not mentioned) (12,13). Adults coded as deaf in both ears were considered to have "severe bilateral hearing impairment." Consistent with the standard reporting approach used by the NCHS, participants falling into categories 2 through 4 were grouped into a single category of "some hearing impairment." For age-group analyses, some hearing impairment and severe bilateral hearing impairment were combined into a single category: "any hearing impairment."
Analyses
Because of the complex sample survey design, all analyses were completed using the Software for Statistical Analysis of Correlated Data (SUDAAN) package to take into account sample weights and design effects (14). To allow for comparisons of hearing impairment over time, we adjusted rates for age using the direct adjustment method using the 2000 population distribution as the standard (15). To determine if there were any statistically significant trends in hearing impairment over the 10-year period, a weighted linear regression model was fitted to the annual adjusted rates. The weight used for each rate was the inverse of its variance. Both linear and higher order polynomial models were examined. Because of the large number of models considered, only those trends with corresponding p values <.01 were considered statistically significant.
|
RESULTS |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
|
|
|
|
DISCUSSION |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
From other studies, there is some limited evidence that hearing rates were increasing in the years prior to the 1986–1994 NHIS. In the Alameda County Study, age-adjusted rates of self-reported hearing loss in adults aged 50 years and older increased from 9.2% in 1965 to 14.0% in 1974, and remained virtually unchanged as shown in their 1983 reassessment (14.1%) (16). In supplemental hearing loss surveys conducted as part of the 1971, 1977, 1990, and 1991 NHIS, age-adjusted rates of any reported trouble with hearing in children and adults were similar in 1971 and 1977 (6.9% and 7.0%) but then increased to 8.6% in the 1990–1991 surveys (17). These findings and those reported for the Alameda County Study are not directly comparable to the results presented in Figure 1 due to differences in age adjustment and the assessment and coding of hearing impairments. However, these findings raise the possibility that hearing impairment rates were increasing in the two decades prior to the 1986–1994 NHIS. If true, then an alternate interpretation of the relatively stable trends reported in Figure 2 is that hearing rates have now stabilized and, if the compression of morbidity hypothesis is applicable, future hearing loss surveys will eventually provide evidence that rates in older adults are declining. Unfortunately, data to test this hypothesis will not be readily forthcoming, as the NHIS was extensively redesigned in 1997, including new hearing loss questions. The NCHS cautions that a direct comparison of prevalence rates of hearing impairment should not be made with data collected in years prior to 1997 (18). Thus, it will be a number of years before trends in hearing impairment can be reliably made using more recently collected NHIS data.
There was no evidence that rates of reported hearing impairment were increasing in young adults, possibly as a result of environmental noise exposure (Figure 2). Initial noise-induced hearing loss typically occurs in the higher frequencies less essential for comprehension of speech (e.g., 3000–6000 Hz) and losses in these higher frequencies may not be perceived by young adults as impaired hearing. The prevalence of impairment in the range of 3000–6000 Hz was 12.5% among National Health and Nutrition Examination Survey (NHANES) participants aged 6–19 years (8), raising the possibility that future generations will experience higher rates of hearing impairment. However, despite reports of excessive environmental noise exposure in adolescents (7), there exist no data supporting the contention that there have been population-level increases in noise exposure over time. Furthermore, clear associations between environmental noise exposures, aside from occupational and firearm exposures, have not been clearly established in any age group (7,19). Nevertheless, findings from the NHANES suggest that longitudinal studies of adolescents are needed to determine if early noise-induced changes in hearing acuity lead to permanent, and perhaps progressive, hearing impairment in young adulthood.
There are several advantages to the use of the NHIS to examine trends in reported hearing impairment. The NHIS is designed to be representative of the U.S. population; only institutionalized and military groups have been omitted from direct sampling. Annual survey response rates were excellent (i.e., 95%–98%). In addition, the collection of proxy information on household participants meant that adults with severe hearing impairments were less likely to be missed, which is a potential problem in telephone-based health surveys (20,21).
The major limitation of the present analyses is the self-reported or proxy-reported nature of ascertainment of hearing impairment in the NHIS. The sensitivity and specificity of self-reported measures of hearing impairment range from 56% to 93% and from 56% to 82%, respectively, when using pure tone audiometric findings as the "gold standard" (22). Pure tone audiometry is perhaps the single best clinical indicator of hearing impairment. However, this clinical procedure is just one of a battery of tests administered by audiologists to quantify the type (e.g., conductive, sensorineural) and degree of hearing impairment as well as to determine if hearing amplification should be prescribed. There is a growing consensus that a battery of clinical and self-reported functional measures are needed to adequately assess hearing impairment (23–25), but that self-reported measures of hearing impairment have acceptable validity for use in epidemiologic studies and in primary care settings in both men and women as well as in younger and older adults (22,26–30). It should be noted, however, that virtually all of these studies were conducted in white populations. Additional research is therefore needed to confirm that similar measures of hearing impairment are also valid for use in African-American populations.
Summary
Analyses of the NHIS data show that annual age-adjusted rates of reported hearing impairment varied only slightly from 1986 to 1995. Also, there were no significant increases or decreases in reported hearing impairment rates in any of the 10-year age groups. Population-based studies designed to include clinical and self-reported measures of hearing impairment are needed to further examine trends in hearing impairment.
|
Acknowledgments |
|---|
Received October 15, 2003
Accepted March 25, 2004
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. L. Lam, D. J. Lee, O. Gomez-Marin, D. D. Zheng, and A. J. Caban Concurrent Visual and Hearing Impairment and Risk of Mortality: The National Health Interview Survey Arch Ophthalmol, January 1, 2006; 124(1): 95 - 101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Caban, D. J. Lee, O. Gomez-Marin, B. L. Lam, and D. D. Zheng Prevalence of Concurrent Hearing and Visual Impairment in US Adults: The National Health Interview Survey, 1997-2002 Am J Public Health, November 1, 2005; 95(11): 1940 - 1942. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
