Tracking population health based on self-reported impairments: trends in the prevalence of hearing loss in US adults, 1976-2006

Published May 18, 2009, in American Journal of Epidemiology (opens in a new window)

Abstract

Research shows that Americans are hearing better today than they were 30 years ago, but progress on reducing hearing loss has slowed. The study, Tracking population health based on self-reported impairments: trends in the prevalence of hearing loss in US adults, 1976-2006, is the first to provide comparable estimates of long-term trends in the prevalence of hearing loss among US adults from the late 1970s to the early 2000s. The work was done in collaboration with researchers at the Harvard School of Public Health.

Research findings

The researchers found that the rate of hearing loss decreased dramatically in the 1990s, from a high in 1992 of about 12.2% of all men and 7% of all women, to a low of 8.1% and 4.2%, respectively, in 2000. Since then, hearing loss has remained about the same, indicating that efforts to prevent hearing loss through noise reduction and better hearing protection in high-noise work places have hit a wall. 
 
Previous studies have shown wide variations in the prevalence of hearing loss. In 1999, for example, the hearing loss among men was estimated at 25.9% in one survey, 16.2% in another, and 7.5% in a third. This study found that such discrepancies in reported hearing loss can be attributed to the different wording and response scales of survey instruments. Additionally, biases in self-reports may vary systematically with age (i.e., the likelihood of reporting hearing loss attenuates with increasing age). 
 
Although it is considered the gold standard for measuring hearing loss, audiometric testing has been used much less frequently than self-reports, the study found. Out of 990,609 people who participated in hearing loss studies since 1976, only about 5,300 were given the headphone tests. By finding the ways that self-reported measures differ systematically from the audiometric tests, however, the researchers were able to estimate how many people with hearing loss were being missed by the self-reported surveys. 
 
A unique contribution of the present study is to overcome differences in measurement methods across time periods and surveys in order to link these past trends to more recent measurements. The large sample sizes in this study allowed the researchers to use regression methods to make full use of available data and to enable adjustment and comparison of self-reports by leveraging audiometric testing data from more recent years.
 
The study results on survey-specific biases in self-reported hearing status reinforce the need for caution in interpreting and comparing self-reports in different surveys. Nevertheless, the findings indicate that partial data on measured performance may be exploited successfully to correct for these differences. This approach may have general utility for enhancing the comparability of various health impairments and symptoms measured with self-reports and, thus, for improving the validity of monitoring long-term trends in population health.

Analytical approach

Data were obtained from interviews and examinations in four nationally representative survey series conducted between 1976 and 2006 in the United States. Measured data were obtained from the pure-tone air conduction audiometry performed in three cycles of the National Health and Nutrition Examination Survey (NHANES). Although it is considered the gold standard for measuring hearing loss, audiometric testing has been used much less frequently than self-reports, the study found. Out of 990,609 people who participated in hearing loss studies since 1976, only about 5,300 were given the headphone tests.
 
The prevalence of measured or self-reported hearing loss was computed from individual surveys by sex and year. Next, unit record data of all surveys were merged to investigate differences in self-reports due to variation in question wording across surveys. The authors performed logistic regression to quantify self-reporting biases compared with audiometric measurements. Trends in age-standardized prevalence of bilateral hearing loss were estimated with corrections for self-reporting biases. 

Research objective

Trends in the prevalence of hearing loss among US adults remain ambiguous because of variation across surveys in question wording and limited use of audiometric examinations. In this study, researchers used partial data on audiometric testing to examine and correct for biases in self-reports on hearing and develop valid estimates of levels and trends in the prevalence of bilateral hearing loss among US adults from 1976 to 2006. This research is part of ongoing work conducted by IHME to provide timely, accurate, and comparable health measurements.

Recommendations for future work

To understand why the trend in hearing loss reduction has flattened, the researchers propose that workplace regulations in the 1990s reduced exposure to hazardous levels of noise and that further study of regulations and prevention programs may be needed to continue the downward trend.
 
Hearing loss remains an important public health problem in the United States. Following the decline in the prevalence of bilateral hearing loss at the speech frequencies during the 1990s, more recent trends have been stable in both adult men and women. Further strategies to promote prevention of hearing loss at work and in communities may be needed to resume progress in reducing the burden of hearing loss. 
Read full article (opens in a new window)

Citation

Ikeda N, Murray CJL, Salomon JA. Tracking population health based on self-reported impairments: trends in the prevalence of hearing loss in US adults, 1976-2006. American Journal of Epidemiology. 2009 May 18; 170(1):80–87.