Sage Journals: Discover world-class research

Abstract

Over 430 million people worldwide experience disabling hearing loss, a condition that becomes more prevalent with age. Although the genetic component to hearing loss has been well established, there has been less data available regarding changes in the genetic contributions to hearing loss over time. We report the pure tone hearing thresholds across 500, 1,000, 2,000, 4,000, and 8,000 Hz from over 1,000 male twins comprising monozygotic (MZ) and dizygotic (DZ) pairs sampled from the United States-based Vietnam Era Twin Study of Aging (VETSA). Twins were tested during three waves, at an average age of 56 at wave 1, an average age of 62 at wave 2, and an average age of 68 at wave 3. Genetically informed structural equation models were used to calculate the genetic contributions. Genetic factors accounted for between 49.4% and 67.7% of the variance in hearing acuity for all frequencies at all three time points. There was no substantial change in the ratio of genetic versus environmental contributions across the three time points, or across individual acoustic frequencies. The stability of hearing acuity over time was moderate to highly attributable to genetic factors. Change in hearing acuity was better explained by unique person-specific environmental factors. These results, from the largest-scale twin study of hearing acuity to date, replicate previous findings that hearing acuity in late life is significantly determined by genetic factors. The unique contribution of the present analysis is that the proportion of hearing acuity attributed to genetics remains relatively consistent across 12 years.

Keywords

heritability presbycusis longitudinal aging twin study

Introduction

Over 430 million people worldwide have disabling hearing loss, with over 25% of older adults affected (World Health Organization, 2024). If not addressed, even a mild to moderate hearing loss can disrupt communication, diminish life satisfaction, and potentially accelerate age-associated cognitive decline (Gurgel et al., 2014; Humes et al., 2013; Lin, 2011; Lin et al., 2011). In fact, it has been suggested that hearing loss may be a modifiable risk factor for dementia (Lin et al., 2023; Livingston et al., 2024). Although a decline in hearing acuity is one of the hallmarks of adult aging, there is wide variability in hearing among adults of equivalent chronological age, a consequence of a complex mix of genetic and environmental influences (Fransen et al., 2003).

A number of studies have employed twin designs to estimate the relative contributions of genetic and environmental factors to age-related hearing loss. In these studies, the estimation of genetic contributions is based on similarity in hearing acuity between monozygotic (MZ) twin pairs versus the similarity in hearing acuity between dizygotic (DZ) twin pairs. These studies leverage the fact that MZ twins share 100% of their genes while DZ twins share on average 50% of their genes. Such studies have shown genetic factors to account for 40% to 70% of the variation in age-related hearing loss (e.g., Cherny et al., 2020; Duan et al., 2019; Garringer et al., 2006; Karlsson et al., 1997; Viljanen et al., 2007; Wingfield et al., 2007; Wolber et al., 2012).

Although valuable, these studies were limited by cross-sectional designs, offering a single snapshot of genetic and environmental contributions to hearing acuity at one moment in time. An important study, and to our knowledge the first longitudinal twin study of adult age-related hearing loss, was conducted by Bogo et al. (2015) on a Swedish sample of 128 MZ twin pairs and 111 DZ twin pairs. The subjects ranged from age 34 to 79 at Time 1, with the group's hearing re-tested 18 years later. Averaging pure-tone frequencies into ‘low’ versus ‘high’ frequency ranges, they found that genetic factors accounted for 53%–65% of the variance in hearing acuity for both low and high frequencies, figures that did not change appreciably between the initial testing session and the retest 18 years later. By contrast, environmental factors were more impactful for the degree of change in hearing acuity from Time 1 to Time 2, accounting for 77–88% of the variance in hearing acuity in the better ear.

Given that changes in hearing acuity are likely to differ substantially beginning in the third versus the eighth decade of life, the wide age range in Bogo and colleagues’ (2015) study that spanned young adulthood to elderly subjects raises the question of whether a more focused age sample would show analogous results. Of special interest is whether the pattern of genetic and environmental influences on age-related hearing loss is sufficiently robust as to appear in the critical time span between the mid-fifties, when for many individuals, hearing first appears as a problem (Cruickshanks et al., 1998), and early old age.

As part of the United States-based Vietnam Era Twin Study of Aging (VETSA; Kremen et al., 2006, 2019) we tested pure tone thresholds across the major speech frequencies for over 1,000 participants, all of whom were members of MZ or DZ twin pairs, at three time points averaging six years apart, beginning with participants at an average age of 56. We wished to determine 1) the change in hearing acuity with age across the speech frequency range, 2) whether the relative contributions of genetic and environmental factors to hearing acuity across three testing points changes or remains stable, 3) whether we may observe a differential effect of genetic and environmental factors on the higher ranges of speech frequencies, as it is the higher frequencies that tend to be more sensitive to aging and environmental insult than the lower frequencies (Baraldi et al., 2015; Vaden et al., 2017), and finally 4) whether the change in hearing acuity over the, on average, 12-year span of this study would be more attributable to genetic or environmental influences.

Methods

Participants

Participants were community dwelling men in the ongoing longitudinal Vietnam Era Twin Study of Aging (VETSA) (Kremen et al., 2019).The participants in this study were randomly recruited from the Harvard Drug Study of male twins from the Vietnam Era Twin Registry (Tsuang et al., 2001). The Registry is a twin registry comprising male twins who had both served in the United States military at some time between 1965 and 1975 (Eisen et al., 1987; Henderson et al., 1990). Participants in the Harvard Drug Study were not selected on the basis of substance use or any diagnostic criteria. VETSA 1 eligibility included being between 51 and 59 years of age at the time of enrollment and both members of a twin pair agreeing to participate.

There were 1,237 total participants during wave 1 in the VETSA database, 1,207 participants during wave 2, and 1,205 participants at wave 3. A subset of participants in waves 2 and 3 were attrition replacements that were demographically matched to returnees. Participants had an average age of 56 (51–60) at wave 1, 62 (56–66) at wave 2 and 68 (61–70) at wave 3. In general, the participants are a reasonably representative community-dwelling sample of U.S. men in their age cohort in terms of sociodemographic as well as health, education, and lifestyle characteristics (Kremen et al., 2006; Schoenborn & Heyman, 2009). Approximately 80% of these participants reported that they did not have combat exposure. For approximately 96% of the participants zygosity was determined through genome-wide genotyping or analysis of 25 microsatellite markers. When these measures were unavailable, zygosity assessment was conducted using validated questionnaires and blood group testing (see Eisen et al., 1987).

Participants were asked if they experience any hearing difficulty, and if they wear hearing aids. There were 536 participants who reported difficulty hearing at wave 1 (43.33%), 505 reported difficulties hearing at wave 2 (41.83%), and 547 at wave 3 (45.39%). There were 44 who reported that they wore hearing aids at wave 1 (3.56%), 107 at wave 2 (8.86%), and 200 at wave 3 (16.60%). It is important to note, however, that only 12 participants were wearing hearing aids on the day of the assessment at wave 1 (0.97%), 62 at wave 2 (5.14%), and 144 at wave 3 (11.95%).

Audiological Assessment

Participants were tested for pure tone hearing acuity using standard audiometric procedures (Harrell, 2002) with a Maico M41 Audiometer (Maico Diagnostics, Eden Prairie, MN). Testing was conducted without hearing aids. Hearing was tested at frequencies of 500, 1,000, 2,000, 4,000, and 8,000 Hz during each of the three waves.

Estimating Heritability

In conventional ACE twin models, latent variables are used to estimate the distinct components of phenotypic variance. Additive genetic variance (A) captures the cumulative effect of individual genes; common or shared environmental variance (C) quantifies the environmental factors shared by individuals that contribute to similarities in traits; and unique environmental variance (E) accounts for the environmental influences that contribute to differences in traits. The proportion of variance due to each factor (A, C and E) can be calculated by dividing the respective path coefficient for A, C, and E from the total variance in the trait.

The proportion of A, C, and E that contribute to and may be shared by multiple traits can be investigated using an extension of this approach. In the present study, hearing thresholds were measured at multiple acoustic frequencies (500, 1,000, 2,000, 4,000, and 8,000 Hz) across three time points. By using a cross-lagged ACE model, the contributions of A, C, and E to contemporaneous correlations (associations between frequency thresholds within the same time point), stability paths (longitudinal associations connecting hearing acuity in the same acoustic frequency over time), and cross-lagged relationships (reciprocal longitudinal associations connecting different acoustic frequencies) can be estimated within a single model (see Malanchini et al., 2017 for a discussion of the cross-lagged ACE model and matrix algebra). Figure 1 shows the path diagram for the initial cross-lagged ACE model.

Figure 1.

Path diagram for the cross-lagged ACE model. Hearing thresholds tested at 1,000, 2,000, 4,000, and 8,000 Hz are denoted as i and interconnected paths are included between these frequencies in the same manner as the paths shown to 500 Hz. Correlations between observed variables within the same time point are also included. The path diagram is shown for one twin.

The cross-lagged ACE model was fit using the TwinAnalysis Package (Voronin, 2024), which fits the cross-lagged model using the structural equation package OpenMx (Neale et al., 2016) for R (R Core Team, 2023). The initial model was reduced by testing nested models where parameters were constrained to zero using the umx package (Bates et al., 2019). Parameters were only restricted if the reduced model did not have significantly worse model fit as indicated by likelihood ratio tests.

For the cross-lagged ACE model, all participants (including attrition replacements) were included even if they only were present for a single wave. Datapoints for specific frequencies within an individual were not included if the threshold was unachieved, resulting in variable participants for each frequency and wave. The models used in this study were maximum likelihood based to account for missing data. Supplementary Table 1 shows the final number of available subjects for each zygosity, wave, and acoustic frequency after invalid and unachieved observations were removed. The intraclass correlations are also shown for each hearing threshold measured.

We also wished to determine the relative contributions of genetic and environmental factors to the presence or degree of change in hearing acuity within this sample. For this purpose, the measure of change was calculated as the slope constant of the better ear PTA (the average of 500, 1,000, 2,000, and 4,000 Hz) across the three waves. Participants who did not participate in at least two of the three waves were excluded, leaving 673 MZ and 493 DZ individual twins for this analysis. A univariate ACE model was used to examine heritability of the degree of change. This model was reduced by testing nested models, and nonsignificant parameters were constrained to zero. For both the univariate and the cross-lagged ACE models, bootstrapped confidence intervals were obtained, and the percent of variance attributable to A, C, and E were calculated using functions from the TwinAnalysis package (Voronin, 2024).

Results

Figure 2 shows mean hearing thresholds at each frequency tested for each of the three testing sessions for the MZ (left panel) and DZ (right panel) twins plotted in the form of an audiogram. One sees the gentle sloping loss across the frequency range typical of age-related hearing loss (presbycusis). As also might be expected, the increasing loss over time is more marked for the higher than the lower frequencies.

Figure 2.

Mean hearing thresholds for monozygotic (left) and dizygotic (right) twins during wave 1 (red), wave 2 (black) and wave 3 (blue). Error bars are one standard error of the mean and in most cases are too small to be visible.

These data were analyzed using linear mixed effects model comparisons using likelihood ratio tests to identify significant effects (LMEMs; Bates, 2010). Fixed effects were tested for Wave (times 1, 2, and 3), Frequency, the interaction of Wave and Frequency, and Zygosity (MZ or DZ). Random effects were included for participant and twin pair. The analysis confirmed a significant effect of Frequency (χ²[1] = 916.16, p < 0.001) and Wave (χ²[1] = 8.41, p < 0.001), with a significant interaction between Wave and Frequency (χ²[1] = 275.59, p < 0.001). There was no significant effect of Zygosity on hearing acuity (χ²[1] = 3.22, p = 0.073).

Figure 3 (top) shows the percent of variance explained by A, C, and E for each acoustic frequency threshold in each wave in the form of a heatmap, where darker colors indicate a greater contribution. From visual inspection, one can see that the percent of variance in hearing acuity explained by additive genetic factors remained relatively stable at around 50% or more. The majority of the remaining variance was explained by unique, person-specific environmental variance, with relatively low contributions from shared environmental variance.

Figure 3.

(top) shows the percentage of variance explained by additive genetic (A), common environmental (C), and shared environmental (E) influences detected for each wave and acoustic frequency in the form of a heatmap. (bottom) displays the amount of raw variance with the total raw variance in red, additive genetic variance (A) in black, common environmental variance (C) in blue, and unique environmental variance (E) in yellow.

Figure 3 (bottom) displays the amount of raw variance that was used for the heritability estimations. One can see an overall pattern of greater variance for the higher frequencies, and that the contributions of genetic variance appear to diverge from the contributions of unique environmental variance at acoustic frequencies tested over 500 Hz.

The lowest specific estimate for additive genetic factors was 49.44% (CI = 38.46% – 62.27%) appearing during wave 2 for 500 Hz while the highest heritability estimate was 67.71% (CI = 60.59% – 71.27%) appearing during wave 3 for 1,000 Hz. Averaging across waves, heritability estimates were lowest for 500 Hz at 51.10% and highest for 1,000 Hz at 60.45%. Averaging across acoustic frequency, heritability estimates were similar during wave 1 (54.70%), wave 2 (55.41%), and wave 3 (58.04%).

The relative contribution of common environmental factors was minimal in the present study. The lowest estimate of C was 1.78% (CI = 0.74% - 8.11%) for 1,000 Hz during wave 1, while the highest estimate of C was 12.22% (CI = 4.80% - 18.67%) for 2,000 Hz during wave 2.

Unique, person-specific environmental factors accounted for, on average, 37.45% of the variance in hearing acuity across all waves and acoustic frequencies. The lowest specific estimate of E was 28.23% (CI = 19.92% - 34.19%) for 1,000 Hz at wave 3, while the highest estimate of E was 43.73% (CI = 23.22% - 51.79%) for 500 Hz at wave 3.

Genetic Contributions to Longitudinal Associations

Hearing thresholds in each wave were predictive of hearing thresholds at the same frequency in the subsequent wave. Supporting this notion, Pearson correlation coefficients between hearing thresholds within the same twin and same acoustic frequency from wave 1 to wave 2, and from wave 2 to wave 3, were higher than .57 for 500 Hz and greater than .74 for all other frequencies tested. These correlations were higher from wave 2 to wave 3 than from wave 1 to wave 2 (see Supplementary Table 2).

Table 1 shows the percent of A, C, and E for the stability paths (associations connecting hearing acuity in the same acoustic frequency over time), as well as the cross-twin cross-trait correlations within each acoustic frequency tested. Cross-twin cross-trait correlations were consistently higher for MZ than for DZ twins for all acoustic frequencies, indicating the presence of genetic effects.

Table 1.

Cross-Twin Cross-Trait Correlations for Longitudinal Associations and the Percent of Variance Attributable to Additive Genetic (A), Common Environmental (C) and Unique Environmental (E) Influences.

Frequency	Wave 1 to Wave 2					Wave 2 to Wave 3
Frequency	rMZ₁	rDZ₂	%A₃	%C₄	%E₅	rMZ₆	rDZ₇	%A₈	%C₉	%E₁₀
500 Hz	0.29	0.16	56.4 (36.46–76.43)	10.23 (0.81–17.15)	33.37 (17.77–48.37)	0.38	0.05	56.13 (40.25–72.58)	4.55 (0.60–24.88)	39.32 (23.27–55.21)
1,000 Hz	0.51	0.21	62.35 (47.79–71.97)	3.69 (0.05–17.36)	33.96 (23.30–41.96)	0.56	0.28	71.69 (55.45–81.69)	10.36 (3.88–28.52)	17.95 (10.22–29.06)
2,000 Hz	0.52	0.17	53.01 (44.21–67.06)	7.93 (1.46–12.67)	39.06 (33.13–47.31)	0.59	0.27	54.84 (37.04–59.87)	5.10 (3.76–10.67)	40.07 (31.48–57.34)
4,000 Hz	0.61	0.29	68.53 (55.80–77.14)	3.78 (0.08–11.45)	27.69 (16.58–37.31)	0.70	0.31	57.06 (26.59–69.75)	14.18 (0.71–24.85)	28.76 (19.99–44.70)
8,000 Hz	0.55	0.14	62.19 (55.52–74.14)	1.25 (0.15–4.22)	36.56 (19.98–41.59)	0.58	0.23	56.81 (41.53–70.76)	5.36 (0.21–13.33)	37.83 (25.45–43.09)

Notes:

1. Cross-twin cross-trait correlation between MZ twin pairs between wave one and wave two.

2. Cross-twin cross-trait correlation between DZ twin pairs between wave one and wave two.

3. Additive genetic component of the variance from the path from wave one to wave two.

4. Common environmental (shared) component of the variance of the path from wave one to wave two.

5. Unique environmental (individual-specific) component of the variance of the path from wave one to wave two.

6. Cross twin cross trait correlation between MZ twin pairs between wave two and wave three.

7. Cross twin cross trait correlation between DZ twin pairs between wave two and wave three.

8. Additive genetic component of the variance from the path from wave two to wave three.

9. Common environmental (shared) component of the variance of the path from wave two to wave three.

10. Unique environmental (individual-specific) component of the variance of the path from wave two to wave three.

95% confidence intervals are shown within parenthesis.

As seen in Table 1, the variance in the stability of hearing acuity was largely determined by genetic factors. For wave 1 to wave 2, specific estimates of the heritability of the stability of hearing loss ranged from 53.01% to 68.53% and from wave 2 to wave 3 ranged from 54.84% to 71.69%. Common environmental contributions to the stability of hearing acuity (shared environmental contributions between twins) were once again minimal, at 14.18% or less. The person-specific unique environmental variance ranged from 27.69% to 40.07%.

The cross-lagged model in this study enables the examination of the relative contributions of A, C, and E to reciprocal relationships between different frequencies over time. In general, there was wide variability in the genetic contributions to the cross-lagged associations. From wave 1 to wave 2, in middle frequencies from 500-2,000 Hz, genetic influence accounted for on average 41.39% of the variance of the cross-lagged associations. Genetic factors were more prominent for explaining variability in cross lagged associations starting at 4,000 and 8,000 Hz, explaining on average 69.20% of the variance. A similar trend was seen for cross lagged links from wave 2 to wave 3, with the genetic component of cross-lagged associations from 500-2,000 Hz explaining 30.89% of the variance on average. In contrast, genetic influence explained 53.47% of the variance of cross-lagged relationships between 4,000-8,000 Hz on average.

Genetic and Environmental Contributions to Degree of Change

As mentioned in the methods section, we also wished to determine the genetic contribution to the degree of change in hearing acuity over time within this sample. Similar to the other phenotypes in this study, the intraclass correlations for the degree of change over time were higher for MZ twins than for DZ twins at .42 and .15, respectively. For the degree of change analysis, the more parsimonious AE model was chosen after model comparison. The variance in the degree of change in hearing acuity accounted for by genetic effects was 40.40% (CI = 30.17% - 57.15%) while the variance explained by unique, person-specific effects accounted for 59.60% of the total variance (CI = 42.85% - 69.83%).

Discussion

The present data showed for both MZ and DZ twins a decline in hearing with age to a differentially greater effect in the high frequency ranges. This pattern is representative of men in this age group based on population studies (Cruickshanks et al., 1998). A number of studies comparing the hearing acuity of adult MZ and DZ twins conducted at a single time point have reliably demonstrated a significant contribution of genetic factors to the hearing acuity of middle aged and older adults (Duan et al., 2019; Garringer et al., 2006; Karlsson et al., 1997; Viljanen et al., 2007; Wingfield et al., 2007; Wolber et al., 2012). Less has been known about the genetic contributions to hearing acuity over time as the individual ages. Our average 12-year longitudinal study has demonstrated the relative genetic contribution to hearing acuity (approximately 50% or more of the variance) to remain stable at three tested time points as individuals moved from middle age to early older adulthood. This extends the findings of a smaller scale, two time-point longitudinal study conducted on a Swedish sample that, to our knowledge, was the sole longitudinal twin study of adult hearing in the extant literature as previously noted (Bogo et al., 2015). With an age range of 45 years, it was difficult to know if change in genetic influences over time might differ in young and older adults. The present study showed that the extent of genetic influence on hearing acuity was stable from approximately the mid-50s to the late 60s.

Because the sensitivity to sounds at higher acoustic frequencies tends to show a greater relative decline than lower frequencies in adult aging (Baraldi et al., 2015; Vaden et al., 2017, and our data), we examined whether the genetic contribution may differ for different acoustic frequencies. We examined genetic contributions for each individual frequency within the major speech range (500, 1,000, 2,000, 4,000 and 8,000 Hz). We found that genetic contributions accounted for 49.4% to 67.7% of the variance for all frequencies at all three time points, consistent with previous estimations of the heritability of hearing acuity. That is, although there was more overall variance to be explained within the higher acoustic frequencies, the relative contributions of genetic, unique, and shared environmental sources were similar across the full frequency range tested.

These results are in line with Bogo and colleagues’ (2015) findings. In their case, rather than examining individual frequencies as done here, Bogo and colleagues compared genetic contributions for the lower frequencies tested (based on the mean of thresholds for 500, 1,000, 2,000, and 4,000 Hz) versus the higher frequencies tested (based on the mean of thresholds for 3,000, 4,000, 6,000, and 8,000 Hz). They found genetic factors to have contributed between 53% to 65% for both the means of the lower and higher pure tone frequencies at baseline and at retest 18 years later.

Similar to other studies of hearing, shared environmental factors had a relatively small effect in our study, contributing 12% or less to hearing acuity at each of the three time points across all of the acoustic frequencies tested (cf., Bogo et al., 2015; Wingfield et al., 2007).

Of special note, our statistical approach enabled us to determine the genetic component of the degree to which an individuals’ hearing acuity at one time point is correlated at each acoustic frequency with their hearing acuity at the subsequent time point (the genetic component of the stability of hearing acuity over time). In this regard, we found that the individual's genetic makeup plays a significant role in determining the stability in hearing acuity over time.

As mentioned, we found a gentle sloping loss across the frequency range typical of age-related hearing loss (presbycusis) from average age 56 to age 68. As also might be expected, the increasing loss over time is more marked for the higher than the lower frequencies. In contrast, the genetic contribution to hearing acuity in each frequency and its relative stability over time, were consistent over time. Both genetic and unique environmental factors contributed to the change or lack of change over time. Unique environmental factors, however, appeared to be somewhat more important in determining the change or lack of change within an individual, accounting for approximately 59.6% of the variance. This suggests that while genetics may provide a stable baseline, environmental influences are critical in driving the presence or degree of change in hearing ability from midlife to early old age.

Population based genome-wide association studies have also addressed the question of the heritability of age related hearing loss, finding heritability estimates that are much lower than are found in typical twin designs (e.g., Hoffmann et al., 2016). Discrepancies between heritability estimated from twin studies and heritability estimated from genome-wide association studies are present in a variety of traits, an active debate in the literature referred to as the “missing heritability problem” (Young, 2019). While genome-wide association studies methods are critical for identifying the underlying genetic architecture of a trait, they may underestimate heritability (Manolio et al., 2009). Conversely, twin studies may overestimate genetic effects due to violations of the equal environments assumption (Felson 2014). Although we cannot say with certainty, it is possible that the precise estimate of the heritability of age related hearing loss lies somewhere between estimates provided by genome-wide association studies and conventional twin study designs.

A major limitation of the present study was the all-male sample. This was due to the all-male composition of the Vietnam Era Twin Registry (Eisen et al., 1987; Henderson et al., 1990), from which the participants in the VETSA project were drawn. Indeed, there are several differences in the prevalence and severity of hearing loss between males and females, as males are more likely to acquire hearing loss (e.g., Homans et al., 2017) and males have a faster rate of threshold increase than females (Kim et al., 2010). The rate of change of hearing acuity can also vary by sex at specific frequencies. For example, in a recent large scale study of demographic differences in hearing acuity, Dillard and colleagues (2024) found that from age 60-69 females tend to have higher rates of change in hearing loss at thresholds of 4000 Hz and higher, while males tend to show higher rates of change at thresholds of 1,000 Hz and 2,000 Hz. While there is work that suggests that the heritability of age related hearing loss in older women is around 75% (Viljanen et al., 2007), longitudinal research is needed to determine if that figure may change over time.

Another limitation of this study is that there may have been selective attrition, in that the men who have more difficulty with hearing may have dropped out of the study due to the long cognitive testing sessions that were a part of the VETSA project. Participants in this study were also largely white and non-Hispanic. We should also emphasize that our measure of hearing acuity was limited to pure tone audiometry. Although pure tone thresholds are central to classifications of hearing acuity in clinical audiology (Harrell, 2002), other measures, such as distortion product otoacoustic emissions (DPOAEs), and tympanometry were not examined. From these data we cannot determine the degree to which the genetic influences demonstrated here reflect vulnerability or resilience to hearing loss. In a similar manner, environmental factors can be negative (noise exposure, unmonitored medications with ototoxic properties) and positive (the use of hearing protection in high noise environments).

The present data, from the largest and most comprehensive longitudinal twin study of age-related hearing loss to date from midlife to early old age, present a strong case for both a significant and stable genetic influences on hearing acuity in mid and later life, and for a substantial unique environmental contribution to the presence or degree of change over time. The fact that genetic influences account for approximately one-half to two-thirds of the variance in hearing acuity in adult aging and that heritability remains consistent over time validates the importance of detecting individuals at high risk for hearing loss. This point has taken on special significance with demonstrations that untreated hearing loss can accelerate age-associated cognitive decline (Lin, 2011; Lin et al., 2011). Although not without contention (Dawes & Munro, 2024), it has been suggested that treating hearing loss (i.e., use of a hearing aid) may reduce the risk of dementia (Lin et al., 2023; Livingston et al., 2024).

Supplemental Material

sj-docx-1-tia-10.1177_23312165251320156 - Supplemental material for Genetic and Environmental Contributions to Age-Related Hearing Loss: Results from a Longitudinal Twin Study

Supplemental material, sj-docx-1-tia-10.1177_23312165251320156 for Genetic and Environmental Contributions to Age-Related Hearing Loss: Results from a Longitudinal Twin Study by Ryan M. O'Leary, Arthur Wingfield, Michael J. Lyons, Carol E. Franz and William S. Kremen in Trends in Hearing

Footnotes

Acknowledgements

We thank the members of the Vietnam Era Twin Registry who have taken part in the VETSA project, as well as their families. Without their contribution this research would not have been possible. The authors would also like to thank Ivan Voronin for his correspondence about the cross-lagged ACE model.

Author Contributions

Ryan M. O’Leary: Writing- Original draft preparation. Formal Analyses, Visualization. Arthur Wingfield: Writing- Original draft preparation, reviewing and editing. Supervision. Michael J. Lyons: Writing- Reviewing and editing. Funding Acquisition, Supervision, Conceptualization. Carol E. Franz: Writing- Reviewing and editing. Funding Acquisition, Supervision, Conceptualization. William S. Kremen: Writing- Reviewing and editing. Funding Acquisition, Supervision, Conceptualization.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The content is the sole responsibility of the authors and does not necessarily represent official views of the National Institute on Aging, National Institutes on Health, or the Veteran's Administration.

National Institutes of Health, (grant number AG018286, AG02982, AG050595, AG055367, AG064955, AG076838).

The VETSA project of which this is part, is supported by National Institutes of Health Grants R01 AG018286, R01 AG050595, AG076838, AG064955, P01 AG055367, and R01 AG02982 from the National Institute on Aging. A.W. gratefully acknowledges support from the Stephen J. Cloobeck Research Fund. The U.S. Department of Veterans Affairs has supplied support for the development and maintenance of the Vietnam Era Twin Registry.

Informed Consent

The study was approved by the Human Subjects Committees of all involved institutions, and all participants gave written informed consent.

ORCID iDs

Arthur Wingfield

Carol E. Franz

Ryan M. O’Leary

Data Availability Statement

Data are available for research access (Vietnam Era Twin Study of Aging [VETSA] Investigators, ) https://psychiatry.ucsd.edu/research/programs-centers/vetsa/index.html

Supplemental Material

Supplemental material for this article is available online.

References

Baraldi

G. S.

de Almeida

L. C.

Borges

A. C. C.

(2015). Hearing loss in aging. Brazilian Journal of Otorhinolaryngology, 73(1), 58–64. https://doi.org/10.1016/S1808-8694(15)31123-X

Bates

D. M.

(2010). Ime4: Mixed-effects modeling with R.

Bates

T. C.

Maes

Neale

M. C.

(2019). Umx: Twin and path-based structural equation modeling in R. Twin Research and Human Genetics, 22(1), 27–41. https://doi.org/10.1017/thg.2019.2

Bogo

Farah

Johnson

A. C.

Karlsson

K. K.

Pedersen

N. L.

Svartengren

Skjönsberg

(2015). The role of genetic factors for hearing deterioration across 20 years: A twin study. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences, 70(5), 647–653. https://doi.org/10.1093/gerona/glu245

Cherny

S. S.

Livshits

Wells

H. R.

Freidin

M. B.

Malkin

Dawson

S. J.

Williams

F. M.

(2020). Self-reported hearing loss questions provide a good measure for genetic studies: A polygenic risk score analysis from UK Biobank. European Journal of Human Genetics, 28(8), 1056–1065. https://doi.org/10.1038/s41431-020-0603-2

Cruickshanks

K. J.

Wiley

T. L.

Tweed

T. S.

Klein

B. E.

Klein

Mares-Perlman

J. A.

Nondahl

D. M.

(1998). Prevalence of hearing loss in older adults in Beaver Dam, Wisconsin: The epidemiology of hearing loss study. American Journal of Epidemiology, 148(9), 879–886. https://doi.org/10.1093/oxfordjournals.aje.a009713

Dawes

Munro

K. J.

(2024). Hearing loss and dementia: Where to from here? Ear and Hearing, 45(3), 529–536. https://doi.org/10.1097/AUD.0000000000001494

Dillard

L. K.

Matthews

L. J.

Maldonado

Simpson

A. N.

Dubno

J. R.

(2024). Demographic factors impact the rate of hearing decline across the adult lifespan. Communications Medicine, 4(1), 171. https://doi.org/10.1038/s43856-024-00593-w

Duan

Zhang

Liang

Tian

Pang

Tan

Qiu

(2019). Heritability of age-related hearing loss in middle-aged and elderly Chinese: A population-based twin study. Ear and Hearing, 40(2), 253–259. https://doi.org/10.1097/AUD.0000000000000610

10.

Eisen

True

Goldberg

Henderson

Robinette

C. D.

(1987). The Vietnam Era Twin (VET) registry: Method of construction. Acta Geneticae medicae et Gemellologiae, 36(1), 61–66. https://doi.org/10.1017/S0001566000004591

11.

Felson

(2014). What can we learn from twin studies? A comprehensive evaluation of the equal environments assumption. Social Science Research, 43, 184–199. https://doi.org/10.1016/j.ssresearch.2013.10.004

12.

Fransen

Lemkens

Van Laer

Van Camp

(2003). Age-related hearing impairment (ARHI): Environmental risk factors and genetic prospects. Experimental Gerontology, 38(4), 353–359. https://doi.org/10.1016/S0531-5565(03)00032-9

13.

Garringer

H. J.

Pankratz

N. D.

Nichols

W. C.

Reed

(2006). Hearing impairment susceptibility in elderly men and the DFNA18 locus. Archives of Otolaryngology–Head & Neck Surgery, 132(5), 506–510. https://doi.org/10.1001/archotol.132.5.506

14.

Gurgel

R. K.

Ward

P. D.

Schwartz

Norton

M. C.

Foster

N. L.

Tschanz

J. T.

(2014). Relationship of hearing loss and dementia: A prospective, population-based study. Otology & Neurotology, 35(5), 775–781. https://doi.org/10.1097/MAO.0000000000000313

15.

Harrell

R. W.

(2002). Pure tone evaluation. In Katz

(Ed.), Handbook of clinical audiology (pp. 71–87). Lippincott Williams & Wilkins.

16.

Henderson

W. G.

Eisen

S. F.

Goldberg

True

W. R.

Barnes

J. E.

Vitek

(1990). The Vietnam Era Twin Registry: A resource for medical research. Public Health Reports, 105(4), 368–373.

17.

Hoffmann

T. J.

Keats

B. J.

Yoshikawa

Schaefer

Risch

Lustig

L. R.

(2016). A large genome-wide association study of age-related hearing impairment using electronic health records. PLoS Genetics, 12(10), e1006371. https://doi.org/10.1371/journal.pgen.1006371

18.

Homans

N. C.

Metselaar

R. M.

Dingemanse

J. G.

van der Schroeff

M. P.

Brocaar

M. P.

Wieringa

M. H.

Baatenburg de Jong

R. J.

Brocaar

M. P.

Goedegebure

(2017). Prevalence of age-related hearing loss, including sex differences, in older adults in a large cohort study. The Laryngoscope, 127(3), 725–730. https://doi.org/10.1002/lary.26150

19.

Humes

L. E.

Busey

T. A.

Craig

Kewley-Port

(2013). Are age-related changes in cognitive function driven by age-related changes in sensory processing? Attention, Perception, & Psychophysics, 75, 508–524. https://doi.org/10.3758/s13414-012-0406-9

20.

Karlsson

K. K.

Harris

J. R.

Svartengren

(1997). Description and primary results from an audiometric study of male twins. Ear and Hearing, 18(2), 114–120. https://doi.org/10.1097/00003446-199704000-00003

21.

Kim

Lim

E. J.

Kim

H. S.

Park

J. H.

Jarng

S. S.

Lee

S. H.

(2010). Sex differences in a cross sectional study of age-related hearing loss in Korean. Clinical and Experimental Otorhinolaryngology, 3(1), 27–31. https://doi.org/10.3342/ceo.2010.3.1.27

22.

Kremen

W. S.

Franz

C. E.

Lyons

M. J.

(2019). Current status of the Vietnam era twin study of aging (VETSA). Twin Research and Human Genetics, 22(6), 783–787. https://doi.org/10.1017/thg.2019.125

23.

Kremen

W. S.

Thompson-Brenner

Leung

Y. J.

Grant

M. D.

Franz

C. E.

Panizzon

M. S.

Eisen

S. A.

Jacobson

K. C.

Boake

Lyons

M. J.

(2006). Genes, environment, and time: The Vietnam Era Twin Study of Aging (VETSA). Twin Research and Human Genetics, 9, 1009–1022. https://doi.org/10.1375/twin.9.6.1009

24.

Lin

F. R.

(2011). Hearing loss and cognition among older adults in the United States. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences, 66(10), 1131–1136. https://doi.org/10.1093/gerona/glr115

25.

Lin

F. R.

Metter

E. J.

O'Brien

R. J.

Resnick

S. M.

Zonderman

A. B.

Ferrucci

(2011). Hearing loss and incident dementia. Archives of Neurology, 68(2), 214–220. https://doi.org/10.1001/archneurol.2010.362

26.

Lin

F. R.

Pike

J. R.

Albert

M. S.

Arnold

Burgard

Chisolm

Coresh

(2023). Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): A multicentre, randomised controlled trial. The Lancet, 402(10404), 786–797. https://doi.org/10.1016/S0140-6736(23)01406-X

27.

Livingston

Huntley

Liu

K. Y.

Costafreda

S. G.

Selbæk

Alladi

Mukadam

(2024). Dementia prevention, intervention, and care: 2024 report of the lancet standing commission. The Lancet, 404(10452), 572–628. https://doi.org/10.1016/S0140-6736(24)01296-0

28.

Malanchini

Wang

Voronin

Schenker

V. J.

Plomin

Petrill

S. A.

Kovas

(2017). Reading self-perceived ability, enjoyment and achievement: A genetically informative study of their reciprocal links over time. Developmental Psychology, 53(4), 698–712. https://doi.org/10.1037/dev0000209

29.

Manolio

T. A.

Collins

F. S.

Cox

N. J.

Goldstein

D. B.

Hindorff

L. A.

Hunter

D. J.

McCarthy

M. I.

Ramos

E. M.

Cardon

L. R.

Chakravarti

Cho

J. H.

Guttmacher

A. E.

Kong

Kruglyak

Mardis

Rotimi

C. N.

Slatkin

Valle

Whittemore

A. S.

Boehnke

Clark

A. G.

Eichler

E. E.

Gibson

Haines

J. L.

Mackay

T. F.

McCarroll

S. A.

Visscher

P. M.

(2009). Finding the missing heritability of complex diseases. Nature, 461(7265), 747–753. https://doi.org/10.1038/nature08494

30.

Neale

M. C.

Hunter

M. D.

Pritikin

J. N.

Zahery

Brick

T. R.

Kirkpatrick

R. M.

Estabrook

Bates

Maes

H. H.

Boker

S. M.

(2016). Openmx 2.0: Extended structural equation and statistical modeling. Psychometrika, 81, 535–549. https://doi.org/10.1007/s11336-014-9435-8

31.

R Core Team (2023). R: A language and environment for statistical computing. R Foundation for Statistical Computing.

32.

Schoenborn

C. A.

Heyman

K. M.

(2009). Health characteristics of adults aged 55 years and over: United States, 2004–2007. National Health Statistics Reports, 16, 1–31. https://doi.org/10.1037/e623972009-001

33.

Tsuang

M. T.

Bar

J. L.

Harley

R. M.

Lyons

M. J.

(2001). The Harvard twin study of substance abuse: What we have learned. Harvard Review of Psychiatry, 9(6), 267–279. https://doi.org/10.1080/10673220127912

34.

Vaden

K. I.

Matthews

L. J.

Eckert

M. A.

Dubno

J. R.

(2017). Longitudinal changes in audiometric phenotypes of age-related hearing loss. Journal of the Association for Research in Otolaryngology, 18, 371–385. https://doi.org/10.1007/s10162-016-0596-2

35.

Vietnam Era Twin Study of Aging (VETSA) Investigators (2025). Vietnam Era Twin Study of Aging (VETSA) dataset. University of California San Diego. https://psychiatry.ucsd.edu/research/programs-centers/vetsa/index.html

36.

Viljanen

Era

Kaprio

Pyykkö

Koskenvuo

Rantanen

(2007). Genetic and environmental influences on hearing in older women. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 62(4), 447–452. https://doi.org/10.1093/gerona/62.4.447

37.

Voronin

(2024). TwinAnalysis (Version 0.3.0) [R package]. GitHub. https://github.com/IvanVoronin/TwinAnalysis.

38.

Wingfield

Panizzon

Grant

M. D.

Toomey

Kremen

W. S.

Franz

C. E.

Jacobson

K. C.

Eisen

S. A.

Lyons

(2007). A twin-study of genetic contributions to hearing acuity in late middle age. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 62(11), 1294–1299. https://doi.org/10.1093/gerona/62.11.1294

39.

Wolber

L. E.

Steves

C. J.

Spector

T. D.

Williams

F. M.

(2012). Hearing ability with age in northern European women: A new web-based approach to genetic studies. PLoS One, 7(4), e35500. https://doi.org/10.1371/journal.pone.0035500

40.

World Health Organization. (2024). Deafness and Hearing Loss. WHO website. https://www.who.int/news-room/fact-sheets/detail/deafness-and-hearing-loss. Updated March 20, 2023. Accessed October 28, 2024.

41.

Young

A. I.

(2019). Solving the missing heritability problem. PLoS Genetics, 15(6), e1008222. https://doi.org/10.1371/journal.pgen.1008222

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB