Abstract
Introduction
Hearing loss is common in ageing populations, but thorough investigation of factors associated with objective hearing loss in otherwise healthy, community-dwelling older individuals is rare. We examined prevalence of age-related hearing loss (ARHL) in healthy, community-dwelling older adults, and determined whether sociodemographic, lifestyle, or health factors associate with hearing thresholds. Audiometry assessment was investigated with self-reports of hearing loss and hearing handicap.
Methods
Australian participants (n = 1,260) of median age 73 years (IQR 71–76) joined ASPirin in Reducing Events in the Elderly (ASPREE)-Hearing, a sub-study of the ASPREE trial with exclusions including cognitive impairment, cardiovascular disease, independence-limiting physical disability, and uncontrolled hypertension. ASPREE collected demographics, anthropometrics, lifestyle, and health data. Audiometry measured better ear pure-tone average (PTA) across four frequencies (0.5–4 kHz) to establish hearing thresholds, categorised as normal or mild, moderate, and severe hearing loss. Questionnaires collected perceived hearing problems and noise exposure.
Results
ARHL prevalence by audiometry was 49.7%, affecting men (59%) more than women (41%). A majority (54.5%) self-reported some hearing problems which mostly aligned with objective assessments; 45.6% self-reported a “little trouble” with hearing, while 35% had objective mild hearing loss; 8.3% reported having a “lot of trouble” hearing, while 13% had moderate hearing loss; and 0.6% reported being “deaf,” while 2% demonstrated severe hearing loss. There was a significant association (p < 0.001) between self-reported hearing handicap and audiometric measures of hearing loss. In multivariate analysis of health, demographics, and lifestyle risk factors, only age, gender (men), and education years (<12) remained associated (p < 0.05) with hearing loss. Hearing thresholds were not associated with smoking, living situation, alcohol use, hypertension, diabetes, or chronic kidney disease.
Conclusion
ARHL robustly assessed by audiometry is common among healthy older Australians with men more likely to have abnormal hearing thresholds than women. Hearing loss was associated with fewer years of formal education, but not with a range of chronic conditions or alcohol use. Self-reported hearing loss correlates well with higher PTA hearing threshold levels in this healthy cohort where prevalence was lower than previously reported for the age group 70+ years. Hearing health education remains an important public health tool for this age. Targeting hearing in older patient health checks could be beneficial to mitigate the cognitive, social, and mental health consequences of ARHL, even if patients do not report a problem or handicap.
Keywords: Age-related changes, Community, Hearing impairment, Hearing loss, Older people, Epidemiology
Introduction
Hearing difficulty is a common complaint among older individuals. Hearing loss can have insidious detrimental effects on an older adult’s quality of life and mood, including communication difficulties in group settings, social withdrawal, embarrassment, and reduced self-esteem [1]. Hearing loss has also been associated with significant functional and cognitive decline and frailty [2–6]. Age-related hearing loss (ARHL) is considered the leading cause of years lived with disability, over the age of 70 years [7].
The World Health Organization defines hearing impairment as a four-frequency average hearing threshold greater than 25 decibels hearing level (dBHL) in the better ear [8]. The recent World Hearing Report 2021 [9] predicts the numbers of those with hearing loss may increase more than 1.5-fold during the next 3 decades, with over 700 million likely to experience a moderate or greater degree of hearing loss by 2050. Mild hearing loss (between 26 dBHL and 40 dBHL) is commonly experienced by 40% of diverse national populations in the 70–75-year age group [4] and can be as frequent as 65% in the late 70s age group [10]. Furthermore, the eighth and ninth decades can be a period of profound deterioration [11]. A hearing threshold of >40 dBHL, or moderate impairment, is accompanied by a significant burden of disability [12]. In Australia, data drawn from the Dynamic Analyses to Optimise Ageing Project (DYNOPTA) (n = 50,652) showed that the prevalence of hearing impairment increases by 13.5% each additional 5 years over 50 years of age [10].
ARHL or presbycusis is a multifactorial and gradual loss of hearing in both ears with age, often affecting people over the age of 60 years. The prevalence of ARHL may depend on physiological risk factors, genetic predisposition, accumulated environmental noise exposure, and health comorbidities such as cardiovascular disease (CVD) [13–16]. Associations have been suggested by some studies, although not universally shown, with hypertension, diabetes, and smoking [17–19].
Re-evaluation of previously acknowledged risk factors [15, 17–19] for hearing loss in independent-living older adults is worthwhile, especially in the changing landscape of age-related illness. The ASPirin in Reducing Events in the Elderly (ASPREE) clinical trial, conducted in Australia and the USA between 2010 and 2017 with main results from the trial published in 2018 [20–22], provided the opportunity to study the impact of ARHL in an otherwise healthy subgroup of Australian participants. We aimed to determine the prevalence of hearing loss by audiometric assessment and self-report in ASPREE participants in Australia, aged 70+ years who were enrolled in the ASPREE-Hearing sub-study [23]. The associations between hearing loss and demographics, lifestyle factors, and non-life-threatening chronic health conditions were assessed in this cohort.
Methods
Australian data from ASPREE [20–22], version 3 longitudinal dataset, and its ASPREE-Hearing sub-study [23] were accessed for the present analysis. The ASPREE-Hearing sub-study collected direct hearing measures through audiometry and self-reports of hearing quality through questionnaires, at the baseline and at years 1.5 and 3 of follow-up. The present cross-sectional study uses data from the baseline period only.
The ASPREE Clinical Trial
ASPREE was a multi-centre, double-blind, randomised, placebo-controlled, primary prevention clinical trial designed to determine whether 100 mg enteric-coated daily oral aspirin extended the duration of disability-free survival in 19,114 healthy, community-dwelling participants over a median 4.7 years. The trial was conducted in Australia (n = 16,703) and the USA (n = 2,411), enrolling participants who were at least 70 years of age (US minorities were at least 65 years) [24] between 2010 and 2014. The trial concluded in 2017. The main ASPREE trial results were published in 2018 [20–22].
Eligibility for the ASPREE study depended on informed consent and excluded those with diagnosed conditions such as CVD, dementia, a Modified Mini-Mental State Examination (3MS) [25] score ≤77, physical disability in performing any one of six basic activities of daily living (Katz ADLs), high risk of bleeding, any condition considered potentially fatal within 5 years, uncontrolled high blood pressure, and those prescribed aspirin for ongoing conditions [24, 26]. At baseline, ASPREE participants completed questionnaires (demographics, lifestyle, medical history, mood, depression, quality of life, basic activities of daily living), and physiological and anthropometric assessments were conducted. Details of these measures and other aspects of the ASPREE clinical trial have been previously published [24, 26].
Other Assessments and Measures from ASPREE
The following information, relevant to the present analyses, was collected through the ASPREE parent study by questionnaire at enrolment: age, living circumstances, education history, health behaviours including categories of alcohol or cigarette consumption (former, current, and never), prescription medication, and medical history. Body weight, waist circumference, and height were measured in person.
Chronic kidney disease (CKD) was defined as eGFR below 60 mL/min/1.73 m2, or albumin-to-creatinine ratio ≥3 mg/mmol. Hypertension was defined as the mean of three blood pressure readings ≥140 systolic or ≥90 diastolic or on treatment for high blood pressure. Diabetes mellitus was defined through self-report of diabetes and/or high fasting blood glucose level ≥126 mg/dL or 7 mmol/L and/or treatment with diabetes medications.
The ASPREE-Hearing Sub-Study
Hearing Sub-Study Participant Eligibility
During the final year of recruitment into the ASPREE parent study, new Australian ASPREE participants enrolled between April and December 2014 were also eligible to participate in the ASPREE-Hearing sub-study [23], unless they had bilateral cochlear implants or deep ear canal implant hearing aids. Written informed consent was obtained from ASPREE-Hearing participants. The study protocol was approved by the Human Research Ethics Committee of Monash University (CF14/920–2014000376).
Objective Hearing Assessments: Pure-Tone Audiometry
All hearing assessments, in addition to other ASPREE data collection, were performed at baseline of the ASPREE study by trained staff, according to a standardised study operating procedure [23], designed with support from the Hearing Cooperative Research Centre in Melbourne, Australia. Staff were trained for adherence to the protocol in conducting mobile audiometry assessments and were regularly monitored for compliance with the standard operating procedures.
Hearing tests were conducted without hearing aids, and otoscopy assessment for ear obstruction(s) was performed prior to audiometry [23]. Any previously unknown hearing impairment detected during this sub-study was conveyed to the participant’s general practitioner, and a copy of the audiogram provided, if the pure-tone average (PTA) was outside the normal range, i.e., ≥26 dBHL.
Standardised pure-tone audiometry was conducted using portable Interacoustics AD226 audiometers (Interacoustics audiometers, Allé, Denmark, distributed through North Ryde, NSW, Australia) with ER3A fitted earphone inserts and sound-attenuating earmuffs (Australian Standards AS/NZ 1270:2002). This method reduced the potential impact of ambient noise allowing routine audiometrics to be conducted outside sound-proof rooms or specialist hearing centres [27]. Tests were conducted in general practice clinics, research centre facilities, or mobile research vehicles (with the engine turned off) and the vehicle located in a quiet area [23].
Hearing thresholds were recorded at six standard audiometric frequencies. Pure-tone audiometry was used to obtain air conduction thresholds, in both ears wherever possible, at six standard octave frequencies for hearing thresholds (0.25, 0.5, 1, 2, 4, and 8 kHz). A PTA of air conduction thresholds at 0.5, 1, 2, and 4 kHz in the better ear is the summary measure of hearing ability used in these analyses. Where a presented tone frequency at any level did not elicit a response from the participant, the maximum possible presentation level plus 5 dB was given as the sound detection threshold (105 dBHL + 5 dB = 110 dBHL). A hearing threshold beyond the normal range of 0–25 dBHL is considered a hearing problem [7]. The better ear hearing threshold PTAs (continuously scored) were categorised into four levels: normal (0–25 dBHL), mild (26–40 dBHL), moderate (41–60 dBHL), or severe hearing loss (>60 dBHL).
Subjective Hearing Assessments: Hearing-Related Questionnaires
Hearing sub-study participants completed the short 10-item Hearing Handicap Inventory for the Elderly-S (HHIE-S) questionnaire of perceived hearing handicap to describe experience without hearing aids [28], included as online supplementary Table S1 (for all online suppl. material, see https://doi.org/10.1159/000541895). This inventory generates a score out of 40 where each question is scored out of 4: “no problems” = 0 points, “sometimes” = 2 points, and “often” = 4 points. The ten responses together are scored out of 40 points. Participants also completed an 11-item questionnaire previously administered in the Baltimore Longitudinal Study of Aging BLSA [29], included as online supplementary Table S2, for an overview of self-reported general hearing ability (without hearing aids), use of hearing aids, and historical noise exposure.
Statistical Analysis
The statistical methods used were t tests, one-way ANOVA tests, simple linear regression, and, finally, a multivariate regression. Simple linear regression was initially applied to examine associations with single factors (listed under Other Assessments and Measures) selected from likely correlates according to the literature [30]. Of note, CVD risk factors were included rather than CVD events which were an exclusion criterion for the ASPREE clinical trial. In addition to potential risk factors indicated by the prior literature, those characteristics with p ≤ 0.1 significance were applied in a multivariate linear regression analysis model, adjusting for age, gender, education, living status, hypertension, smoking, diabetes, and CKD.
Results
Baseline Characteristics
The baseline characteristics of the ASPREE-Hearing study participants are provided in Table 1 and compared with the Australian ASPREE participants who were not in the hearing sub-study. Characteristics of the hearing subgroup were very similar to those not enrolled in the hearing sub-study. Minor differences in the hearing subgroup included a smaller proportion of women (52% vs. 56%) and slightly younger age (by 1 year).
Table 1.
Characteristics of ASPREE-Hearing participants and Australian ASPREE participants who were not involved in the sub-study
| Variable | Categories | ASPREE-Hearing participants (n = 1,260), n (%) | Australian ASPREE participants not enrolled in ASPREE-Hearing (n = 15,443), n (%) |
|---|---|---|---|
| Age, mean (SD), years | 74.4 (4.1) | 75.4 (4.4) | |
| Age groups | 70–74 years | 859 (68.8) | 8,810 (57.1) |
| 75–79 years | 263 (20.9) | 4,168 (27.0) | |
| 80+ years | 138 (11.0) | 2,465 (16.0) | |
| Gender | Men | 603 (47.9) | 6,921 (44.8) |
| Women | 657 (52.1) | 8,522 (55.2) | |
| Education | ≤12 years | 708 (56.2) | 10,247 (57.4) |
| >12 years | 552 (43.8) | 7,606 (42.6) | |
| Living arrangementsa | Alone | 370 (29.4) | 4,961 (32.1) |
| Living with others | 890 (70.3) | 10,482 (67.9) | |
| Smoking status | Current | 40 (3.2) | 521 (3.4) |
| Former | 518 (41.1) | 6,334 (41.0) | |
| Never | 702 (55.7) | 8,588 (55.6) | |
| Alcohol status | Current | 1,034 (82.1) | 12,149 (78.5) |
| Former | 59 (4.7) | 755 (4.9) | |
| Never | 167 (13.3) | 2,539 (16.5) | |
| Ethnicityb | Not Hispanic | 1,246 (98.9) | 15,341 (99.4) |
| Hispanic/Latino | 14 (1.1) | 101 (0.7) | |
| Racec | Caucasian | 1,244 (98.8) | 15,227 (98.7) |
| Non-Caucasian | 15 (1.2) | 208 (1.3) | |
| Handedness | Right | 1,153 (91.5) | 16,378 (91.7) |
| Left | 107 (8.5) | 1,472 (8.3) | |
| Hypertensiond | 916 (72.7) | 11,609 (75.2) | |
| CKDe | 241 (21.0) | 3,818 (26.7) | |
| Polypharmacyf | 324 (25.7) | 4,027 (26.1) | |
| Diabetesg | 148 (11.8) | 1,496 (9.7) | |
| BMI (kg/m2) | <20 underweight | 15 (0.1) | 298 (0.2) |
| 20–24.9 normal | 281 (22.4) | 3,783 (24.6) | |
| 25.0–29.9 overweight | 570 (45.5) | 6,913 (45) | |
| ≥30.0 obese | 399 (31.8) | 4,366 (28.4) | |
| Waist, cm | 98.4 (12.6) | 96.9 (12.7) |
Less than 0.09% of all variables were missing.
aLiving arrangements: “living with others” category includes with family/friends/spouse/communal.
bEthnicity was self-report collected as Hispanic/Latino or non-Hispanic/Latino.
cRace included Caucasian, Aboriginal/Torres Strait Islanders, Hawaiian/other Pacific Islander/Maori, Asian, American Indian, Black/African American, or other.
dHypertension, blood pressure ≥140 systolic or ≥90 diastolic or on treatment for high blood pressure.
eCKD, eGFR ≤60 mL/min/1.73 m2 or UACR ≥3 mg/mmol.
fPolypharmacy ≥5 prescribed medications.
gDiabetes mellitus, self-report of diabetes and/or high fasting blood glucose level ≥126 mg/dL and/or treatment with diabetes medications.
Audiometry Measures
Of the 16,703 Australian ASPREE participants, 2516 were given the opportunity to join the ASPREE-Hearing sub-study, and a subset of 1,270 participants consented to participate (online suppl. Fig. S1). Ten were excluded from this analysis because they did not complete the audiometry activities during the visit. Of the 1,260 active participants, all were able to provide a PTA result in at least one ear, with 11 who were initially assessed as having bilateral obstruction and who returned for a reassessment after treatment. Among the 1,260 participants, 51 had an obstructed right ear canal and 21 had an obstructed left ear canal. In both scenarios, the obstructed ear was not examined further during the test attendance. The better ear descriptor was applied to the ear with available data if only one ear was tested. Some participants had no hearing response to the maximum presentation level of 105 dBHL in at least one of the six stimulus tone frequencies (137 events).
The PTA hearing threshold levels were approximately normally distributed (online suppl. Fig. S2). The left and right ears showed similar ranges and medians, so no further sub-analysis for differences was explored. Across the sample (n = 1,260), the mean hearing threshold PTA was 27.6 dBHL (standard deviation 12.70 dBHL), with minimum value of 5 dBHL and maximum of 82.5 dBHL. Normal PTA (≤25 dBHL) was measured in half (50.3%) the participants with the category prevalences of hearing loss of 34.7% (mild), 13.3% (moderate), and 1.7% (severe).
A box plot of the PTA according to age groups is visualised in Figure 1, demonstrating higher PTA median, 25th and 75th centiles with each successively older age group, in 5-year age brackets from 70 years. Figure 1b also shows a higher PTA median and centiles in men than women. Further, within each of the categories of hearing loss (normal/mild/moderate/severe), in Figure 1c, the older cohort have higher PTA hearing thresholds in each category than the younger.
Fig. 1.
Box plots of four-frequency PTA dBHL in the better ear according to age groups and by gender (n = 1,260). a Hearing level according to age groups. b Hearing level according to gender. c Hearing level according to age groups and hearing problem groups. d Hearing level according to gender and hearing problem groups. The horizontal lines in each graph have been marked at 25 decibels to illustrate the limit between normal hearing (below line) and mild/moderate/severe hearing loss (above line). Box limits represent Q1 (or 25th centile) and Q3 (or 75th centile), with mid-line in each box representing Q2 (or 50th centile). Whiskers represent Q1– 1.5 × IQR and Q3+ 1.5 × IQR. Dots represent outliers >Q3+ 1.5 × IQR. IQR, interquartile range.
Men have higher hearing thresholds than women in every category except for severe hearing loss, though the number of women in this category was very low (Fig. 1d). The relationships, according to one-way ANOVA between PTA and demographics, lifestyle factors, or chronic conditions, are shown in Table 2. PTAs in the better ear worsened with age by approximately 5 dBHL for each age bracket: 70–74 years, 75–79 years, and >80 years (p < 0.001 for differences between the groups). Men had a higher hearing threshold PTA than women (p < 0.001), while fewer formal education years were associated with higher PTA hearing threshold (p < 0.001). The box plot in online supplementary Figure S3 shows median hearing thresholds differed across 6 levels of formal education from <9 to 17+ years with the lowest better ear PTA in those with +17 years of education. On univariate analysis in Table 2, hearing thresholds were significantly associated with living arrangements (p = 0.03), smoking status (p = 0.03), and CKD (p = 0.03) but not associated with race, ethnicity, alcohol intake, hypertension, diabetes, handedness, waist circumference, or BMI. Due to the low numbers of non-Caucasians in the study (15 participants) and non-Hispanic/Latinos (14), no further conclusions could be drawn about race or ethnicity.
Table 2.
Relationship between hearing threshold levels of the better ear PTA and demographic characteristics, lifestyle, and chronic conditions
| Variable | Categories | Better ear PTA threshold decibels | |
|---|---|---|---|
| mean (SD) | p value | ||
| Age | 70–74 years (n = 859) | 25.3 (11.9) | |
| 75–79 years (n = 263) | 30.5 (12.9) | <0.001 | |
| 80+ years (n = 138) | 36.7 (13.3) | ||
| Gender | Men (n = 603) | 29.3 (12.3) | |
| Women (n = 657) | 26.1 (13.2) | <0.001 | |
| Education | ≤12 years (n = 708) | 28.8 (13.3) | |
| >12 years (n = 552) | 26.2 (12.1) | <0.001 | |
| Living arrangementsa | Alone (n = 370) | 28.9 (13.5) | |
| Living with others (n = 890) | 27.2 (12.6) | 0.03 | |
| Smoking status | Ever smoked (n = 558) | 28.5 (12.8) | |
| Never smoked (n = 702) | 26.9 (13.0) | 0.03 | |
| Alcohol status | Current (n = 1,034) | 27.6 (12.6) | |
| Former (n = 59) | 28.7 (14.1) | ||
| Never (n = 167) | 27.8 (13.8) | 0.81 | |
| Ethnicityb | Not Hispanic (1,246) | 27.6 (12.9) | |
| Hispanic/Latino (n = 14) | 28.7 (12.3) | 0.76 | |
| Racec | Caucasian (n = 1,244) | 27.6 (12.8) | 0.40 |
| Non-Caucasian (n = 15) | 28.7 (12.3) | ||
| Handedness | Right hand (n = 1,153) | 27.6 (12.7) | |
| Left hand (n = 107) | 28.1 (14.6) | 0.73 | |
| Hypertensiond | Yes (n = 344) | 27.8 (12.7) | |
| No (n = 916) | 27.2 (13.2) | 0.46 | |
| CKDe | Yes (n = 241) | 29.2 (12.9) | |
| No (n = 907) | 27.3 (12.8) | ||
| Missing (n = 112) | 26.8 (12.5) | 0.04 | |
| Polypharmacyf | Yes (n = 324) | 27.9 (13.1) | |
| No (n = 936) | 27.6 (12.8) | 0.72 | |
| Diabetesg | Yes (n = 148) | 28.2 (12.0) | |
| No (n = 1,112) | 27.6 (13.0) | 0.62 | |
| BMI (kg/m2) | <20 (n = 15) | 30.6 (15.8) | |
| 20–24.9 normal (n = 270) | 27.4 (13.5) | ||
| 25–29.9 overweight (n = 570) | 27.9 (12.6) | ||
| ≥ 30.0 obese (n = 399) | 27.2 (12.8) | 0.63 | |
Less than 0.09% of all variables were missing.
aLiving arrangements: “living with others” category includes with family/friends/spouse/communal.
bEthnicity was self-report collected as Hispanic/Latino or non-Hispanic/Latino.
cRace included Caucasian, Aboriginal/Torres Strait Islanders, Hawaiian/other Pacific Islander/Maori, Asian, American Indian, Black/African American, or other.
dHypertension, blood pressure ≥140 systolic or ≥90 diastolic or on treatment for high blood pressure.
eCKD, eGFR ≤60 mL/min/1.73 m2 or UACR ≥3 mg/mmol.
fPolypharmacy ≥5 prescribed medications.
gDiabetes mellitus, self-report of diabetes and/or high fasting blood glucose level ≥126 mg/dL and/or treatment with diabetes medications.
In multivariate analysis (Table 3), only older age, being male, and having lower education remained significantly associated with hearing threshold (p values <0.001, <0.001 and 0.01 respectively). In the adjusted model, smoking was negatively related to hearing threshold but did not reach statistical significance (p = 0.13, and a confidence interval −0.86 to 0.1).
Table 3.
Multiple linear regression of the relationship between exposures and hearing threshold PTAs
| Variable | β (S.E.) | p value | 95% CI |
|---|---|---|---|
| Age | 0.91 (0.09) | <0.001 | 0.73–1.09 |
| Gender | |||
| Men | Reference | ||
| Women | −3.36 (0.76) | <0.001 | −4.85 to −1.87 |
| Education | |||
| ≤12 years | Reference | ||
| >12 years | −0.46 (0.18) | 0.01 | −0.82 to −0.10 |
| Living arrangementsa | |||
| Alone | Reference | ||
| Living with others | −0.68 (0.42) | 0.10 | −1.5 to 0.13 |
| Smoking statusb | |||
| Ever smoked | Reference | ||
| Never smoked | −0.37 (0.25) | 0.13 | −0.86 to 0.15 |
| Hypertensionc | |||
| Yes | Reference | ||
| No | −0.39 (0.89) | 0.63 | −1.99 to 1.22 |
| CKDd | |||
| Yes | Reference | ||
| No | −0.29 (0.91) | 0.74 | −2.06 to 1.49 |
| Diabetese | |||
| Yes | Reference | ||
| No | −0.13 (1.1) | 0.91 | −2.3 to 2.03 |
aLiving arrangements: “living with others” category includes with family/friends/spouse/communal.
bSmoking status: “ever smoked” category is a combination of former and current smokers.
cHypertension, blood pressure ≥140 systolic or ≥90 diastolic, or on treatment for high blood pressure.
dCKD, eGFR ≤60 mL/min/1.73 m2 or UACR ≥3 mg/mmol.
eDiabetes mellitus, self-report of diabetes and/or high fasting blood glucose level ≥126 mg/dL and/or treatment with diabetes medication.
Self-Reported Hearing Questions
Overall, 1,260 participants completed the HHIE-S questionnaire with frequency of responses shown in online supplementary Table S1. The total score from the HHIE-S is reported in relation to the PTA hearing loss categories, represented as box plots in Figure 2. High-score HHIE-S reflects greater perceived hearing handicap. An HHIE-S score of 8 has been used as a limit as per previous studies where ≤8 is considered no handicap [31].
Fig. 2.
Box plots of HHIE-S total scores among four hearing loss categories. Box limits represent Q1 (or 25th centile) and Q3 (or 75th centile), with mid-line in each box representing Q2 (or 50th centile). Whiskers represent Q1− 1.5 × IQR and Q3+ 1.5 × IQR; dots = outliers >Q3+ 1.5 × IQR. Normal hearing category is ≤25 dBHL PTA hearing level; mild 26–40 dBHL; moderate 41–60 dBHL; severe >60 dBHL. HHIE-S score distribution was asymmetric and positively skewed, so median and percentiles are reported. IQR, interquartile range.
Difficulty “hearing whispers” was reported (sometimes or often) by 66.5% of respondents and difficulty in a restaurant was found in 50.6%, while 22.5% reported “feeling handicapped” by a hearing problem. HHIE-S total scores were lowest among those who demonstrated normal hearing thresholds (0–25 dBHL) and highest in those with severe hearing loss with a threshold >60 dBHL. Two-way ANOVA confirmed a strongly significant (p < 0.001) association between HHIE-S score and categories of hearing loss.
Of note, outliers also occurred in the normal hearing group, with some reporting high HHIE-S scores, as shown in Figure 2. Of those in the normal hearing group, 11% claimed a hearing handicap (high HHIE-S score >8), with frequency more among men (15%) than women (9%). Of those in the moderate or severe hearing loss groups, 5% (43 respondents) claimed no hearing handicap (low HHIE-S score ≤8) with little difference between men (5.2%) and women (4.9%).
Of the 1,241 respondents who completed the BLSA hearing questionnaire on how best to describe their hearing without any hearing aids, 565 or 45.5% described “good” hearing, 566 or 45.6% reported a “little trouble” with hearing, 103 (8.3%) described having a “lot of trouble” hearing, while 7 (0.6%) described themselves as “deaf” without a hearing aid. A plot of individual responses to this question versus PTA in dBHL (online suppl. Fig. S4) illustrates the close relationship between individual answers to this question and actual hearing threshold with increasing median and 25th and 75th centiles with self-report of increasing problems.
A small number of the respondents (2.4%) had experienced ear surgery, although the timing or nature of the surgery was not specified. A history of possible noise exposure through outside work was reported by 175 or 14% of 1,250 respondents, with 82% of these being men. Further, a history of noise exposure from previous use of firearms was reported by 335 or 27% (with 93% being men), exposure of loud noise for 5+ h a week (as a job or other) by 370 or 30% of respondents (84% being men). The total proportion of participants who reported a history of noise exposure of any kind was 526 or 42%. Less than 20% of the cohort used one or more hearing aids (n = 237, 19%), while the majority of those wearing hearing aids (98%) did so bilaterally. Of those wearing a hearing aid, 38% had mild hearing loss, 49% had moderate, and 8% had severe hearing loss in the better ear.
Discussion
The prevalence of hearing loss was 49.7% in a cohort of 1,260 otherwise healthy Australians aged 70+ years, who were enrolled in a hearing sub-study of the ASPREE clinical trial. The descriptor “healthy” is applied on the basis that all participants were free of overt CVD, independence-limiting physical disability, very high blood pressure, or dementia at study enrolment and were deemed by their general practitioner to not have an illness that was likely to be life ending within 5 years [20–22]. The Lancet Global Burden of Disease study of 2019 estimated that the prevalence of any ARHL in individuals aged 75 years is one-third, rising to two-thirds at 85 years [7]. This also aligns with the World Report of hearing loss which shows the global prevalence of moderate or worse hearing loss to be 26% at 70–74 years and 43.6% in the 80–84 years age group [9], with more hearing loss in men than women at each age bracket. The comparable findings in our study are much lower for moderate or worse hearing loss groups (10% at 70–74 years and 20% at 75–79 years), but the increasing trend with age in these international reports including “all comers” [7, 9, 10, 32, 33] is largely consistent with the findings of our study, despite it being conducted in a population with a very low burden of major life-threatening disease.
The strong association of hearing loss with men is consistent with previous literature [9, 32, 33]. We found that in every hearing loss group, even in the normal hearing category, men have worse hearing than women.
In addition to age and gender, increased prevalence of ARHL was associated with fewer years of formal education, consistent with previous findings [31, 32]. Of those with <12 years of formal education, 375 (53%) had some hearing deficit, with 16% showing moderate or severe hearing loss. Fewer years of formal education may relate to occupational noise exposure, although less than one-third of the ASPREE participants reported a history of such exposure.
Unlike findings from other studies [30, 34–36], we did not see an increased prevalence of hearing loss associated with smoking, living status, or the chronic conditions of diabetes mellitus, hypertension, CKD, or obesity, nor any altered association, beneficial or otherwise, with alcohol consumption. CKD was associated with poor hearing thresholds in univariate analysis but not in the multivariate analysis, after accounting for age, gender, and education (Table 3). It is possible that the low percentage of current smokers in our cohort (3%) limits the strength of conclusions about the influence of smoking.
Despite a systematic review providing evidence that alcohol consumption is positively associated with hearing loss [36], in our healthy older ASPREE participants, audiometric assessment of ARHL did not show this association. A possible explanation may be that low-moderate alcohol intake does not adversely affect healthy older adults and most of the ASPREE participants were in the current use category of intake [24]. Of the 18 studies included in the review, there were only two studies with mean participant age older than 65 years and both showed a beneficial effect of moderate alcohol intake [36]. Our study findings of no association with alcohol conflict with the previous older age study set in Australia [37]. Regarding the lack of association of ARHL with the CVD risk factors of diabetes and hypertension, some other studies have reported associations only with diabetes or with neither hypertension nor diabetes [32] or with both risk factors. The absence of clinical manifestation of CVD in our participants may have contributed to a null association, along with other stronger risk factors such as age [32].
Self-reported hearing handicap level, as captured by HHIE-S scores, associated with objectively measured hearing loss (normal to severe) in a level-dependent relationship. This is particularly evident for higher degrees of hearing loss, with stepwise increases in median total HHIE-S scores, as shown in Figure 2. The higher number of outliers in the normal hearing threshold group could be due to the manifestation of central hearing processing problems. Those who claimed a hearing handicap, even though their audiometry did not show a deficit (11%), may be affected by a central processing disorder which is not detectable by audiometry. HHIE-S scores do not discriminate between the better ear and the alternate ear, so perceived hearing handicap may also be related to the alternate ear. Gender differences, higher in men, in reporting hearing loss in the absence of deficit by audiometry were demonstrated. This could be explained by a higher prevalence in men of hearing loss in the alternate ear. There may also be gender difference in the experience or acknowledgement of handicap. Within the HHIE-S questionnaire, the responses to the single question on difficulty hearing in a restaurant most closely approximated the objective audiometry findings of a hearing loss. Despite this association, 77.5% of respondents never felt “handicapped by a hearing problem.” Although the total scores in the HHIE-S concord with hearing threshold, these responses to individual questions suggest that perception of hearing handicap lags behind the disease burden in this healthy older population, confirming earlier findings [38] of under-reporting or misreporting, particularly increasing with age. Nevertheless, these findings continue to support the inventory tool utility, especially if access to audiometry may be limited, or patients are unwilling to undertake formal audiometry testing.
The self-report of hearing within the BLSA questionnaire showed a prevalence (54.5%) of perceived hearing problems. Hearing aid prevalence of 19% was much lower than the prevalence of any self-reported hearing problems, self-reported hearing handicap score >8, or objective measures of hearing loss, although it is not known why hearing aid intervention was not taken up by more of these health-aware participants in a clinical trial. However, among those who self-reported a lot of hearing trouble or deafness (8.9%) the prevalence of hearing aid use is much higher. Likely, obstacles to hearing aid use in the mild hearing loss group include the high cost, stigma associated with the wearing of aids, lack of accessibility to sensitive healthcare professionals, poor customer satisfaction due to poor fit and discomfort, and perception that hearing loss does not affect health [39].
The main limitation of our study is that it is cross-sectional, with no causal inference possible, and only reports of associations can be made, along with prevalence data. Further limitations include a bias towards healthy volunteers, limiting the generalisability of the findings. Nevertheless, characteristics of the ASPREE-Hearing sub-study participants were very similar to the remainder of the ASPREE cohort (Table 1) which was broadly generalisable to the greater Australian population, except for reporting less chronic disease than anticipated for age [24]. This sampling limitation could potentially underestimate hearing loss prevalence in the wider population, but it does represent the healthy, community-based, independent-living older population that is steadily growing in Australia and elsewhere. The lack of bone conduction data in this study restricts further analysis of any association with noise exposure.
Strengths of note include that our findings were taken from older adults living in metropolitan and rural areas, and with differing education backgrounds. Rather than recruitment from a centralised hearing clinic, hearing tests were made possible in many settings with the use of portable audiometry equipment to assess the participants near their home. Objective and subjective measures were collected allowing for comparisons of both data sources. As the hearing study was embedded in a larger clinical trial, access to high quality and rigorously captured non-hearing health data were also an advantage.
In conclusion, this study confirms the high prevalence of hearing loss in an Australian older cohort participating in a primary prevention clinical trial. Otherwise healthy, older adults have a high prevalence of presbycusis which was associated with increasing age, gender (men), and fewer years of formal education. The contribution from other possible risk factors, smoking, diabetes, hypertension, and alcohol consumption, differed from other literature, which is often dominated by younger cohorts. The group with mild-severe hearing loss was predominantly men, those of older age, or those with fewer years of formal education. Self-reported hearing handicap was aligned to audiometric measures of hearing loss in the groups with mild, moderate, or severe hearing loss, supporting the use of hearing assessment by questionnaire where audiometry is not practical or possible. Hearing aid prevalence was higher than the prevalence of self-reports of deafness or a lot of trouble hearing, demonstrating good interest in health for those with more severe hearing problems. The epidemiological insights from this study support efforts to encourage older men and women to explore hearing loss diagnosis and earliest interventions.
Acknowledgments
We thank the participants, staff, and all investigators of ASPREE and ASPREE-Hearing studies. We thank Professor Frank Lin for his guidance.
Statement of Ethics
The ASPREE-Hearing study was approved by the Human Research Ethics Committee of Monash University (CF14/920–2,014,000,376). Written informed consent was obtained prior to enrolment in the study, and the research was conducted in accordance with the World Medical Association Declaration of Helsinki.
Conflict of Interest Statement
The authors have no conflicts of interests to declare.
Funding Sources
The ASPREE clinical trial was supported by grants from the National Institute on Aging and the National Cancer Institute at the National Institutes of Health (U01AG029824 and U19AG062682); the National Health and Medical Research Council of Australia (334047 and 1127060); and by Monash University and the Victorian Cancer Agency (Australia). ASPREE-Hearing was supported by Monash University. The funders had no role in the design, data collection, data analysis, and reporting of this study.
Author Contributions
E.S., J.J.M., R.L.W., G.R., and A.M.M. designed the ASPREE-Hearing study. R.L.W. directed recruitment and operations of the ASPREE-Hearing sub-study, ensuring integration into the ASPREE clinical trial. M.R.N. and N.S. supported the data collection in their respective regions. C.J.B., R.L.W., and J.R. planned this secondary analysis of the ASPREE-Hearing sub-study. C.J.B. wrote the first draft and project-managed the ASPREE-Hearing sub-study. All authors reviewed and approved the final submission.
Funding Statement
The ASPREE clinical trial was supported by grants from the National Institute on Aging and the National Cancer Institute at the National Institutes of Health (U01AG029824 and U19AG062682); the National Health and Medical Research Council of Australia (334047 and 1127060); and by Monash University and the Victorian Cancer Agency (Australia). ASPREE-Hearing was supported by Monash University. The funders had no role in the design, data collection, data analysis, and reporting of this study.
Data Availability Statement
The data that underpin the findings of this study are available in the ASPREE Safe haven repository. Information about data access is available on the ASPREE website https://aspree.org/aus/.
Supplementary Material.
References
- 1. Dalton DS, Cruickshanks KJ, Klein BE, Klein R, Wiley TL, Nondahl DM. The impact of hearing loss on quality of life in older adults. Gerontologist. 2003;43(5):661–8. [DOI] [PubMed] [Google Scholar]
- 2. Loughrey DG, Kelly ME, Kelley GA, Brennan S, Lawlor BA. Association of age-related hearing loss with cognitive function, cognitive impairment, and dementia: a systematic review and meta-analysis. JAMA Otolaryngol Head Neck Surg. 2018;144(2):115–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Brewster KK, Hu M-C, Wall MM, Brown PJ, Zilcha-Mano S, Roose SP, et al. Age-related hearing loss, neuropsychological performance, and incident dementia in older adults. J Alzheimers Dis. 2021;80(2):855–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Panza F, Lozupone M, Sardone R, Battista P, Piccininni M, Dibello V, et al. Sensorial frailty: age-related hearing loss and the risk of cognitive impairment and dementia in later life. Ther Adv Chronic Dis. 2019;10:2040622318811000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Ford AH, Hankey GJ, Yeap BB, Golledge J, Flicker L, Almeida OP. Hearing loss and the risk of dementia in later life. Maturitas. 2018;112:1–11. [DOI] [PubMed] [Google Scholar]
- 6. Tian R, Almeida OP, Jayakody DMP, Ford AH. Association between hearing loss and frailty: a systematic review and meta-analysis. BMC Geriatr. 2021;21(1):333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. GBD 2019 Hearing Loss Collaborators, Kamenov K, Briant PS, Orji AU, Steinmetz JD, Abdoli A. Hearing loss prevalence and years lived with disability, 1990-2019: findings from the Global Burden of Disease Study 2019. Lancet. 2021;397(10278):996–1009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. World Health Organisation . Deafness and hearing loss. 2013. [Google Scholar]
- 9. World report on hearing. Geneva: World Health Organization; 2021. Licence: CC BY-NC-SA 3.0 IGO. [Google Scholar]
- 10. 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):439–58. [DOI] [PubMed] [Google Scholar]
- 11. Kiely KM, Gopinath B, Mitchell P, Luszcz M, Anstey KJ. Cognitive, health, and sociodemographic predictors of longitudinal decline in hearing acuity among older adults. J Gerontol A Biol Sci Med Sci. 2012;67(9):997–1003. [DOI] [PubMed] [Google Scholar]
- 12. Access Economics . Listen Hear!: the economic impact and cost of hearing loss in Australia. Australian Policy Online; 2006. [Google Scholar]
- 13. Wells HRR, Freidin MB, Zainul Abidin FN, Payton A, Dawes P, Munro KJ, et al. GWAS identifies 44 independent associated genomic loci for self-reported adult hearing difficulty in UK biobank. Am J Hum Genet. 2019;105(4):788–802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Nash SD, Cruickshanks KJ, Zhan W, Tsai MY, Klein R, Chappell R, et al. Long-term assessment of systemic inflammation and the cumulative incidence of age-related hearing impairment in the epidemiology of hearing loss study. J Gerontol A Biol Sci Med Sci. 2014;69(2):207–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Engdahl B, Aarhus L, Lie A, Tambs K. Cardiovascular risk factors and hearing loss: the HUNT study. Int J Audiol. 2015;54(12):958–66. [DOI] [PubMed] [Google Scholar]
- 16. Ohlemiller KK. Mechanisms and genes in human strial presbycusis from animal models. Brain Res. 2009;1277:70–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wattamwar K, Qian ZJ, Otter J, Leskowitz MJ, Caruana FF, Siedlecki B, et al. Association of cardiovascular comorbidities with hearing loss in the older old. JAMA Otolaryngol Head Neck Surg. 2018;144(7):623–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Gupta S, Eavey RD, Wang M, Curhan SG, Curhan GC. Type 2 diabetes and the risk of incident hearing loss. Diabetologia. 2019;62(2):281–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Garcia Morales EE, Ting J, Gross AL, Betz JF, Jiang K, Du S, et al. Association of cigarette smoking patterns over 30 years with audiometric hearing impairment and speech-in-noise perception: the Atherosclerosis Risk in Communities Study. JAMA Otolaryngol Head Neck Surg. 2022;148(3):243–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. McNeil JJ, Woods RL, Nelson MR, Reid CM, Kirpach B, Wolfe R, et al. Effect of aspirin on disability-free survival in the healthy elderly. N Engl J Med Overseas Ed. 2018;379(16):1499–508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. McNeil JJ, Nelson MR, Woods RL, Lockery JE, Wolfe R, Reid CM, et al. Effect of aspirin on all-cause mortality in the healthy elderly. N Engl J Med Overseas Ed. 2018;379(16):1519–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. McNeil JJ, Wolfe R, Woods RL, Tonkin AM, Donnan GA, Nelson MR, et al. Effect of aspirin on cardiovascular events and bleeding in the healthy elderly. N Engl J Med Overseas Ed. 2018;379(16):1509–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Lowthian JA, Britt CJ, Rance G, Lin FR, Woods RL, Wolfe R, et al. Slowing the progression of age-related hearing loss: rationale and study design of the ASPIRIN in HEARING, retinal vessels imaging and neurocognition in older generations (ASPREE-HEARING) trial. Contemp Clin Trials. 2016;46:60–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. McNeil JJ, Woods RL, Nelson MR, Murray AM, Reid CM, Kirpach B, et al. Baseline characteristics of participants in the ASPREE (ASPirin in reducing events in the elderly) study. J Gerontol A Biol Sci Med Sci. 2017;72(11):1586–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Teng EL, Chui HC. The modified mini-mental state (3MS) examination. J Clin Psychiatry. 1987;48(8):314–8. [PubMed] [Google Scholar]
- 26. ASPREE Investigator Group . Study design of ASPirin in Reducing Events in the Elderly (ASPREE): a randomized, controlled trial. Contemp Clin Trials. 2013;36(2):555–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Clark JG, Brady M, Earl BR, Scheifele PM, Snyder L, Clark SD. Use of noise cancellation earphones in out-of-booth audiometric evaluations. Int J Audiol. 2017;56(12):989–96. [DOI] [PubMed] [Google Scholar]
- 28. Ventry IM, Weinstein BE. The hearing handicap inventory for the elderly: a new tool. Ear Hear. 1982;3(3):128–34. [DOI] [PubMed] [Google Scholar]
- 29. Baltimore Longitudinal Study of Aging . Self-reported hearing and noise exposure questionnaire.
- 30. Jayakody DMP, Friedland PL, Martins RN, Sohrabi HR. Impact of aging on the auditory system and related cognitive functions: a narrative review. Front Neurosci. 2018;12:125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Sindhusake D, Mitchell P, Smith W, Golding M, Newall P, Hartley D, et al. Validation of self-reported hearing loss. The blue mountains hearing study. Int J Epidemiol. 2001;30(6):1371–8. [DOI] [PubMed] [Google Scholar]
- 32. Lin FR, Thorpe R, Gordon-Salant S, Ferrucci L. Hearing loss prevalence and risk factors among older adults in the United States. J Gerontol A Biol Sci Med Sci. 2011;66(5):582–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Homans NC, Metselaar RM, Dingemanse JG, van der Schroeff MP, Brocaar MP, Wieringa MH, et al. Prevalence of age-related hearing loss, including sex differences, in older adults in a large cohort study. Laryngoscope. 2017;127(3):725–30. [DOI] [PubMed] [Google Scholar]
- 34. Dawes P, Cruickshanks KJ, Moore DR, Edmondson-Jones M, McCormack A, Fortnum H, et al. Cigarette smoking, passive smoking, alcohol consumption, and hearing loss. J Assoc Res Otolaryngol. 2014;15(4):663–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Fransen E, Topsakal V, Hendrickx J-J, Van Laer L, Huyghe JR, Van Eyken E, et al. Occupational noise, smoking, and a high body mass index are risk factors for age-related hearing impairment and moderate alcohol consumption is protective: a European population-based multicenter study. J Assoc Res Otolaryngol. 2008;9(3):264–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Qian P, Zhao Z, Liu S, Xin J, Liu Y, Hao Y, et al. Alcohol as a risk factor for hearing loss: a systematic review and meta-analysis. PLoS One. 2023;18(1):e0280641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Gopinath B, Flood VM, McMahon CM, Burlutsky G, Smith W, Mitchell P. The effects of smoking and alcohol consumption on age-related hearing loss: the Blue Mountains Hearing Study. Ear Hear. 2010;31(2):277–82. [DOI] [PubMed] [Google Scholar]
- 38. Tsimpida D, Kontopantelis E, Ashcroft D, Panagioti M. Comparison of self-reported measures of hearing with an objective audiometric measure in adults in the English Longitudinal Study of Ageing. JAMA Netw Open. 2020;3(8):e2015009–e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. McCormack A, Fortnum H. Why do people fitted with hearing aids not wear them? Int J Audiol. 2013;52(5):360–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that underpin the findings of this study are available in the ASPREE Safe haven repository. Information about data access is available on the ASPREE website https://aspree.org/aus/.