Abstract
This cross-sectional study evaluates 2613 participants 60 years or older in the NHANES study (2001-2012) for variations in self-reported hearing trouble by audiometric hearing loss and sociodemographic characteristics.
The estimated number of adults 60 years or older with hearing loss is anticipated to rise to 62.4 million by 2060.1 In many studies, self-reported hearing trouble is used as a proxy for audiometric measures of hearing. However, self-reported trouble may be influenced by sociodemographic characteristics2 leading to bias in prevalence estimates of hearing loss. How these characteristics may lead to systematic undercounts or overcounts of persons with hearing loss in the population remains unknown. The current study estimates the variations in prevalence and number of people in the United States with self-reported hearing trouble by audiometric hearing and sociodemographic characteristics.
Methods
The analytic sample of this cross-sectional study (Table 1) comprised 2613 adults 60 years or older from the National Health and Nutrition Examination Survey (NHANES; 2001-2012 cycles). Better-ear audiometric hearing was defined according to World Health Organization guidelines3 as normal hearing (≤25 dB hearing level [HL]), mild hearing loss (>25 dB HL but ≤40 dB HL), or moderate or greater hearing loss (>40 dB HL). In NHANES, participants self-reported their hearing as excellent, good, a little trouble, moderate trouble, or deaf. Consistent with previous research,2 responses were converted to a binary measure. Excellent or good responses were categorized as no self-reported trouble, whereas other responses were categorized as self-reported trouble. A sensitivity analysis explored the impact of categorizing a little trouble as no self-reported trouble.
Table 1. Summary Unweighted Sociodemographic Characteristics of the Analytic Sample From NHANES 2001-2012 Cycles.
| Characteristic | Audiometric Hearing, No. (%)a | P Value | ||
|---|---|---|---|---|
| Normal (n = 1562 [59.8%])b | Mild Loss (n = 667 [25.5%])b | Moderate or Greater Loss (n = 384 [14.7%])b | ||
| Female | 892 (57.1) | 287 (43.0) | 115 (29.9) | <.001 |
| Age, y | ||||
| 60-69 | 1060 (67.9) | 243 (36.4) | 89 (23.2) | <.001 |
| 70-79 | 460 (29.4) | 357 (53.5) | 216 (56.3) | |
| 80+ | 42 (2.7) | 67 (10.0) | 79 (20.6) | |
| Race/ethnicity | ||||
| Hispanic | 326 (20.9) | 116 (17.4) | 64 (16.7) | <.001 |
| Non-Hispanic | ||||
| Black | 446 (28.6) | 110 (16.5) | 42 (10.9) | |
| White | 693 (44.4) | 407 (61.0) | 267 (69.5) | |
| Other | 97 (6.2) | 34 (5.1) | 11 (2.9) | |
| Education | ||||
| <High school | 450 (28.8) | 248 (37.2) | 157 (40.9) | <.001 |
| High school | 351 (22.5) | 197 (29.5) | 83 (21.6) | |
| >High school | 761 (48.7) | 222 (33.3) | 143 (37.2) | |
| Unknown | 0 | 0 | 1 (0.3) | |
| Poverty index ratioc | ||||
| <1.25 | 329 (21.1) | 185 (27.7) | 107 (27.9) | <.001 |
| 1.25-3.49 | 622 (39.8) | 282 (42.3) | 172 (44.8) | |
| ≥3.50 | 487 (31.2) | 149 (22.3) | 75 (19.5) | |
| Missing | 124 (7.9) | 51 (7.6) | 30 (7.8) | |
Abbreviations: HL, hearing level; NHANES, National Health and Nutrition Examination Survey.
Percentages may not sum to 100 due to rounding.
Audiometric hearing defined by a 4-frequency (0.5-, 1.0-, 2.0-, and 4.0-kHz) better-ear pure-tone average threshold as normal hearing (≤25 dB HL), mild hearing loss (>25 dB but ≤40 dB HL), or moderate or greater hearing loss (>40 dB HL).
Poverty index ratio is a calculated value in the NHANES data set that is the ratio of family income to poverty; NHANES uses poverty guidelines from the Department of Health and Human Services to calculate the poverty index ratio such that family income is divided by the poverty guidelines according to family size, year, and state. Values greater than 5 were coded in NHANES as 5 owing to disclosure concerns, and as such, poverty index ratio ranges from 0 to 5. Values less than 1 indicate that the participant family income was below the poverty level, and values greater than 1 indicate that the participant family income is above the poverty level.
Sociodemographic variables included age, sex, race/ethnicity, education, and poverty index ratio. Prevalence estimates by sociodemographic characteristics were applied to the US Census Bureau 2018 population estimates.4 To account for the complex sampling design, analytic weights were used. All analyses were conducted in R, version 3.4.1 (R Core Team, 2017).
Results
Although the estimated number of people 60 years or older with self-reported trouble (28.68 million) was similar to the estimated number with audiometric hearing loss (28.35 million), estimates varied by audiometric hearing and sociodemographic characteristics (Table 2). Self-reported hearing trouble underestimated the number of people with audiometric mild hearing loss by 42%. Self-reported trouble undercounted the number of people aged 70 to 79 years and 80 years or older with hearing loss by 15% and 36%, respectively, equivalent to an undercount of 5.1 million people. Self-reported trouble over-counted the number of 60- to 69-year-olds with hearing loss by 35%, equivalent to 3.4 million people. Self-reported trouble overestimated the number of non-Hispanic blacks with hearing loss by 10%. Self-reported trouble undercounted the number of people with hearing loss with low education and low poverty income ratios but overcounted the number of people with hearing loss with high education and high poverty income ratios.
Table 2. Estimated Prevalence and Number of Persons 60 Years or Older in the 2018 US Adult Civilian Population With Audiometric Hearing Loss and Self-reported Hearing Troublea.
| Characteristic | Audiometric Hearing Lossb | Self-reported Hearing Troublec | Difference in No. of Peopled | |||
|---|---|---|---|---|---|---|
| % (95% CI) | No., Millions | % (95% CI) | No., Millions | Relative % Difference | Difference, No., Millions | |
| Overall, >25 dB | 39.84 (36.62-43.07) | 28.35 | 40.31 (37.89-42.73) | 28.68 | 1.16 | 0.33 |
| Audiometric hearinge | ||||||
| Normal | NA | NA | 21.98 (19.55-24.41) | 9.41 | NA | 9.41 |
| Mild | 26.29 (23.45-29.12) | 18.70 | 57.70 (54.06-61.34) | 10.79 | –42.30 | –7.91 |
| Moderate or greater | 13.56 (11.95-15.16) | 9.65 | 87.90 (84.27-91.53) | 8.48 | –12.12 | –1.17 |
| Sex | ||||||
| Female | 32.29 (28.33-36.24) | 12.50 | 31.99 (28.81-35.17) | 12.38 | –0.96 | –0.12 |
| Male | 49.01 (44.53-53.49) | 15.90 | 50.39 (47.07-53.71) | 16.35 | 2.83 | 0.45 |
| Age | ||||||
| 60-69 | 26.17 (22.90-29.43) | 9.70 | 35.22 (32.02-38.43) | 13.06 | 34.64 | 3.36 |
| 70-79 | 54.78 (49.64-59.91) | 12.20 | 46.51 (43.09-49.93) | 10.36 | –15.08 | –1.84 |
| >80 | 78.28 (73.50-83.06) | 9.24 | 50.46 (42.40-58.53) | 5.96 | –35.50 | –3.28 |
| Race/ethnicity | ||||||
| Hispanic | 37.93 (31.97-43.90) | 2.47 | 37.79 (31.71-43.88) | 2.46 | –0.40 | –0.01 |
| Non-Hispanic | ||||||
| Black | 26.33 (21.83-30.84) | 1.82 | 29.06 (24.87-33.25) | 2.01 | 10.44 | 0.19 |
| White | 41.27 (37.50-45.03) | 21.99 | 41.67 (38.79-44.55) | 22.21 | 1.00 | 0.22 |
| Other | 44.92 (33.38-56.47) | 2.00 | 42.41 (28.91-55.91) | 1.88 | –6.00 | –0.12 |
| Education | ||||||
| <High school | 52.94 (47.36-58.51) | 4.68 | 46.69 (42.07-51.31) | 4.13 | –11.75 | –0.55 |
| High School | 45.02 (39.60-50.45) | 10.24 | 40.71 (36.31-45.12) | 9.26 | –9.57 | –0.98 |
| >High school | 31.29 (27.88-34.70) | 12.38 | 37.20 (33.72-40.67) | 14.72 | 18.90 | 2.34 |
| Poverty index ratiof | ||||||
| <1.25 | 52.05 (46.13-57.97) | 5.18 | 41.62 (36.25-47.00) | 4.14 | –20.08 | –1.04 |
| ≥1.25-<3.50 | 44.00 (39.97-48.03) | 11.70 | 43.22 (40.14-46.31) | 11.50 | –1.71 | –0.20 |
| ≥3.50 | 30.19 (25.77-34.60) | 10.45 | 35.47 (30.48-40.46) | 12.27 | 17.42 | 1.82 |
Abbreviations: HL, hearing level; NHANES, National Health and Nutrition Examination Survey.
Adult civilian population (US residents not in the active-duty military); count estimates from the 2018 US Census Bureau Current Population Survey.4
Audiometric hearing loss defined by a four-frequency (0.5-, 1.0-, 2.0-, 4.0-kHz) better-ear pure-tone average threshold >25 dB HL.
Self-reported hearing trouble defined as self-reported general hearing condition as little trouble, moderate trouble, lot of trouble, or deaf.
Difference in the estimated number of adults with audiometric hearing loss and the number of adults with self-reported hearing trouble. A positive value indicates that self-reported hearing trouble overestimates the number of people with hearing loss. A negative value indicates that self-reported hearing trouble underestimates the number of people with hearing loss.
Audiometric hearing defined by a 4-frequency (0.5-, 1.0-, 2.0-, and 4.0-kHz) better-ear pure-tone average threshold as normal hearing (≤25 dB HL), mild hearing loss (>25 dB but ≤40 dB HL), or moderate or greater hearing loss (>40 dB HL).
Poverty index ratio is a calculated value in the NHANES data set that is the ratio of family income to poverty; NHANES uses poverty guidelines from the Department of Health and Human Services to calculate the poverty index ratio such that family income is divided by the poverty guidelines according to family size, year, and state. Values greater than 5 were coded in NHANES as 5 owing to disclosure concerns, and as such, poverty index ratio ranges from 0 to 5. Values less than 1 indicate that the participant family income was below the poverty level, and values greater than 1 indicate that the participant family income is above the poverty level.
When a little trouble hearing was categorized as not troublesome, the estimated number of individuals with self-reported trouble reduced to just 10.16 million, substantially undercounting the number of people with hearing loss.
Discussion
In this nationally representative sample of older adults, the prevalence of self-reported hearing trouble varied across sociodemographic characteristics. Self-reported trouble underestimated the number of people with hearing loss for those 70 years or older, those with a low poverty index ratio, and those with low education. Our findings are consistent with, but also extend, previous research.2
Self-reported hearing trouble is a critical determinant of obtaining hearing aids.5 As hearing aids become available over the counter, there is the possibility that adults who overestimate their hearing trouble may purchase devices unnecessarily and without engaging with hearing care clinicians. Alternatively, individuals who underestimate their hearing trouble who may benefit from devices may not seek them out. However, ongoing efforts to transform the service delivery landscape are addressing accessibility barriers.6 Further research will be needed to evaluate the impact of future over-the-counter hearing aid availability on different sociodemographic groups. The present study is cross-sectional and therefore does not determine the impact of hearing loss duration on self-reported hearing trouble. The number of adults 80 years or older in the analytic sample was relatively small.
Conclusions
In conclusion, self-reported hearing trouble varies by sociodemographic characteristics leading to underestimates of hearing loss in older adults, those with low poverty index ratios, and those with lower education. Therefore, self-reported hearing trouble as a proxy for audiometric hearing testing should be used with caution when estimating hearing loss prevalence among specific subpopulations.
References
- 1.Goman AM, Reed NS, Lin FR. Addressing estimated hearing loss in adults in 2060. JAMA Otolaryngol Head Neck Surg. 2017;143(7):733-734. doi: 10.1001/jamaoto.2016.4642 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kiely KM, Gopinath B, Mitchell P, Browning CJ, Anstey KJ. Evaluating a dichotomized measure of self-reported hearing loss against gold standard audiometry: prevalence estimates and age bias in a pooled national data set. J Aging Health. 2012;24(3):439-458. doi: 10.1177/0898264311425088 [DOI] [PubMed] [Google Scholar]
- 3.World Health Organization Grades of hearing impairment. https://www.who.int/pbd/deafness/hearing_impairment_grades/en/.
- 4.U.S. Census Bureau Current Population Survey, Annual Social and Economic Supplement. https://www.census.gov/cps/data/cpstablecreator.html. Published 2018. Accessed October 26, 2018.
- 5.Ng JH, Loke AY. Determinants of hearing-aid adoption and use among the elderly: a systematic review. Int J Audiol. 2015;54(5):291-300. doi: 10.3109/14992027.2014.966922 [DOI] [PubMed] [Google Scholar]
- 6.Nieman CL, Lin FR. Increasing access to hearing rehabilitation for older adults. Curr Opin Otolaryngol Head Neck Surg. 2017;25(5):342-346. doi: 10.1097/MOO.0000000000000386 [DOI] [PMC free article] [PubMed] [Google Scholar]