Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Oct 12.
Published in final edited form as: Ear Hear. 2022 Mar-Apr;43(2):487–494. doi: 10.1097/AUD.0000000000001111

Impact of hearing aid use on falls and falls-related injury: Results from the Health and Retirement Study

Kristal M Riska 1,2, Sarah B Peskoe 3, Maragatha Kuchibhatla 3, Alexander Gordee 3, Juliessa Pavon 2,4, Se Eun Kim 3, Jessica S West 2,5, Sherri L Smith 1,2,6
PMCID: PMC9554784  NIHMSID: NIHMS1838172  PMID: 34334680

Abstract

Objective:

Falls are considered a significant public health issue and falls risk increases with age. There are many age-related physiologic changes that occur that increase postural instability and the risk for falls (i.e., age-related sensory declines in vision, vestibular, somatosensation, age-related orthopedic changes, and polypharmacy). Hearing loss has been shown to be an independent risk factor for falls. The primary objective of this study was to determine if hearing-aid use modified (reduced) the association between self-reported hearing status and falls or falls-related injury. We hypothesized that hearing aid use would reduce the impact of hearing loss on the odds of falling and falls-related injury. If hearing aid users have reduced odds of falling compared to non-hearing aid users, then that would have an important implications for falls prevention healthcare.

Design:

Longitudinal, survey-based panel study using existing publically-available data from the Health and Retirement Study (HRS).

Results:

A total of 17,923 individuals were grouped based on a self-reported history of falls. Self-reported hearing status was significantly associated with odds of falling and with falls-related injury when controlling for demographic factors and important health characteristics. Hearing aid use was included as an interaction in the fully-adjusted models and the results showed that there was no difference in the association between hearing aid users and non-users for either falls or falls related injury.

Conclusions:

The results of the current study show that when examining self-reported hearing status in a longitudinal sample, hearing aid use does not impact the association between self-reported hearing status and the odds of falls or falls-related injury.

INTRODUCTION

Falls result in increased morbidity, mortality, and healthcare utilization and are the leading cause of accidental death (Florence et al., 2018a; Hartholt et al., 2011; Peel, 2011). In 2014, older Americans experienced 29 million falls causing seven million injuries and costing $50 billion in total falls-related expenditures (Florence et al., 2018). In addition to fractures and other fall-related injuries, falls lead to disability, risk of institutionalization/loss of independence, activity restriction and reduction in quality of life (Davis et al., 2015; Delbaere, Crombez, Vanderstraeten, Willems, & Cambier, 2004; Dunn, Furner, & Miles, 1993; Li, Fisher, Harmer, McAuley, & Wilson, 2003; Stenhagen, Ekström, Nordell, & Elmståhl, 2014; Tinetti & Williams, 1997; Vellas, Cayla, Bocquet, De Pemille, & Albarede, 1987).

It is well established that falls and falls risk increase with increasing age. By 2060, the US Census Bureau estimates that Americans over the age of 65 years will account for nearly a quarter of the US population. As such, the public health impacts of falls are tremendous and of growing concern given the aging population. Falls may be a result of extrinsic and/or intrinsic factors (Dhalwani et al., 2017; Guirguis-Blake, Michael, Perdue, Coppola, & Beil, 2018; Haran et al., 2010; Ivers, Cumming, & Mitchell, 2002; Kulmala et al., 2009; Montero-Odasso et al., 2019; Rubenstein, 2006; Vu, Finch, & Day, 2011). Extrinsic factors include environmental factors such as slippery surfaces, poor stair design, loose rugs/floor mats that may result in a fall. Intrinsic falls risk factors, those internal to the individual, are often multi-factorial in nature. That is, among older adults, there is an increased risk for age-related sensory changes that are critically important for maintaining balance (i.e., age-related vision loss, age-related vestibular loss, and age-related changes to proprioception), and emerging data suggesting that age-related hearing loss also may be important. These age-related sensory changes coupled with other known risk factors for falls (such as polypharmacy and cognitive impairment) leads to multi-factorial disequilibrium being common in older adults, and resulting in increased risk for falls, falls-related injury and loss of independence.

A growing body of evidence also indicates that hearing contributes to balance, and hearing loss and hearing handicap are independently associated with falls in older adults (Deandrea et al., 2010; Ernst et al., 2020, Gopinath, McMahon, Burlutsky, & Mitchell, 2016; Jiam, Li, & Agrawal, 2016; Kulmala et al., 2009; Lin & Ferrucci, 2012; Lopez et al., 2011). In addition to the independent association between hearing loss/handicap and falls, Lin and Ferrucci (2012) demonstrated a dose-dependent risk between hearing loss and falls such that increasing levels of hearing loss, as measured via pure-tone audiometry, were associated with increasing odds of falling. Specifically, they found that for every 10 dB increase in hearing loss there was a 1.6-fold increase in the odds of falling when adjusting for demographics, cardiovascular risk factors, and vestibular function (as assessed using a measure of postural control). This association was even found in those with mild hearing loss. In fact, their data showed that those with mild hearing loss had a three-fold increase in odds of falling.

Because of the known association between hearing loss and falls risk, it is important to determine if traditional treatments for hearing loss, such as a hearing aid use, may modify this risk. To date, there is relatively sparse and conflicting evidence on whether or not hearing aid use modifies the association between hearing loss and falls. Kamil and colleagues (2016) reported no differences in the odds of falling based on amplification status in individuals with moderate or greater hearing loss; whereas, Gopinath (2016) reported an increased odds of falling among hearing aid users with severe self-perceived hearing handicap and concomitant visual impairment (i.e., dual-sensory impairment). No study to date has examined the impacts of hearing aid use of falls and falls-related injury among those with milder hearing impairment status or across the full range of hearing impairment.

In contrast to the research on falls, there is longstanding evidence that audition contributes to balance (Deviterne, Gauchard, Jamet, Vançon, & Perrin, 2005; Dozza, Horak, & Chiari, 2007; Mergner, Maurer, & Peterka, 2003) and there is emerging evidence to suggest that hearing aid use may improve postural control in individuals with hearing loss, which may potentially reduce falls risk (Berge, Nordahl, Aarstad, & Goplen, 2019; Lacerda, Silva, de Tavares Canto, & Cheik, 2012; Mahmoudi et al., 2019; Negahban, Bavarsad Cheshmeh ali, & Nassadj, 2017; Rumalla, Karim, & Hullar, 2015; Shayman, Earhart, & Hullar, 2017; Shayman, Mancini, Weaver, King, & Hullar, 2017; Stevens, Barbour, Gronski, & Hullar, 2016; Vitkovic, Le, Lee, & Clark, 2016; Weaver, Shayman, & Hullar, 2017). A recent systematic review by Ernst and colleagues (2020) was performed on the correlation between hearing aid and cochlear implant use on measures of balance. From their systematic review, they found significant heterogeneity in the study designs, patient groups, and balance assessment tools that made it impossible to conduct a meta-analysis. In an absence of meta-analysis, they qualitatively concluded that among the 10 studies that met their inclusion criteria, the use of hearing aids and cochlear implants appeared to improve spatial-temporal orientation in laboratory-based settings and that long-term follow up may be needed to fully understand the benefit of amplification on postural stability to account for acclimatization and the effects of neural plasticity (Ernst et al 2020).

There have several proposed mechanisms that may explain the association between hearing loss and falls/balance difficulties. These factors include concomitant vestibular dysfunction, increased cognitive load and/or cognitive factors, reduced spatial awareness, and psychosocial factors (Ernst et al 2020, Li & Lindenberger 2002, Jiam et al 2016, Zuniga et al 2012). As seen in Figure 1, we propose a probable mechanistic model that may help to explain the association between hearing loss and falls (our model is adapted from Whitson et al 2018 who described a mechanistic model explaining the association between sensory loss and cognition). Our model posits that hearing loss may indirectly influence falls through multifactorial pathways (in yellow) or may be directly mediated by common etiology (in orange). In many (not all) of the indirect pathways proposed, hearing aids have been shown to be beneficial.

Figure 1.

Figure 1.

Open in a new tab

Schematic of proposed probable mechanistic model explaining the association between hearing loss and falls. The model posits that the association may be explained by either or both direct or indirect pathways.

Thus, the extent to which hearing aid use may modify falls risk and incident falls in individuals with hearing loss remains unclear. To this end, the current study aimed to determine whether or not hearing aid use modifies the odds of falling and falls injury associated with a wide range of self-reported hearing status and in a longitudinal cohort of older adults.

METHODS

This study was approved by the Duke University Intuitional Review Board (Duke Protocol Pro00104449) as an exempt protocol that utilized publically-available data.

Health and Retirement Study (HRS) Overview

The design and administration of the HRS is well-established and reported in detail elsewhere (Staff, 2008). Briefly, HRS is a nationally-representative, longitudinal survey of households with at least one adult over the age of 50 years, is funded by the National Institute on Aging (NIA U01AG009740), and has been under the direction of the University of Michigan since its inception in 1992. Participants are surveyed (either face-to-face or by telephone) every 2 years for the core survey components, resulting in a sample of approximately 20,000 respondents with an overall response rate of greater than 80% (Staff, 2017). The original study was designed to evaluate the transition from active worker into retirement and to examine the interactions between health, family, and economic variables in the retirement period. To that end, the core survey includes questions regarding demographic information, healthcare use, work status, family structure, mental and physical health, finances, and housing (Servais, 2010). De-identified data are available for download from the HRS website (https://hrs.isr.umich.edu/data-products).

The present study used seven cycles of data from the biennial core survey during cycle years 2004-2016 (2004, 2006, 2008, 2010, 2012, 2014, and 2016). Unique identifiers allowed for tracking of individual respondents over the multiple HRS survey cycles. We included individuals 65 years of age and older with data available from at least one of the HRS cycles of interest. The time of entry into our study was defined as the first cycle of data after the individual reached 65 years of age.

Variables of Interest

The hearing-related variables of interest were hearing aid status and self-reported hearing status. Hearing aid use and self-reported hearing status since the previous survey cycle were obtained from two self-report questions. First, participants were asked, “Do you ever wear a hearing aid?” Response options were “yes”, “no”, “don’t know”, or “refused to answer”. Hearing aid users were defined as those individuals indicating “yes” to ever wearing a hearing aid. Second, all participants were asked to rate their hearing (while wearing a hearing aid as usual, if applicable) on a 5-point scale (excellent, very good, good, fair, poor). For both self-reported hearing status and hearing aid use, “refused to answer” and “don’t know” were treated as missing data.

The outcome variables of interest for the analysis were falls and falls injury during each survey cycle. Falls were defined by the question “Have you fallen down in the last two years?” and response options were “yes”, “no”, “don’t know”, or “refused to answer”. Those responding “yes” were categorized as a faller whereas those responding “no” were categorized as a non-faller. For those stating they had fallen in the past 2 years, a follow up question was asked about injury: “In that fall/In any of these falls, did you injure yourself seriously enough to need medical treatment?” and response options were “yes”, “no”, “don’t know”, or “refused to answer”. In both of the falls-related questions, “don’t know” and “refused to answer” were treated as missing data.

Potential confounding variables pertaining to falls and fall-related injury included demographic variables and important health-related factors. Specifically, age was used as the longitudinal time scale, and we extracted the following self-report variables to serve as covariates in our statistical models: age, sex, race, ethnicity, years of education, self-reported eyesight, self-reported memory, depression, smoking status (i.e., history of smoking and current status), history of hypertension, history of diabetes, history of stroke, history of cancer, history of heart disease, self-rated health, self-rated trouble with pain, frequency of vigorous physical activity, history of arthritis, hospitalization in the past two years, and number of activities of daily living (ADLs) performed, trouble with instrumental activities of daily living, and obese BMI. These confounding variables capture the two primary sources of attrition in longitudinal studies of aging populations, namely age and cognitive impairment as it relates to frailty (Chatfield et al, 2005).

Statistical Analysis

A Generalized Estimating Equations (GEE) approach was used to fit logistic regression models (unadjusted, adjusted for demographics and common comorbidity, fully adjusted for demographics, common comorbidity, chronic disease and physical function) for the longitudinal data. GEE is a statistical approach that consistently estimates a population-level effect while accounting for the correlation that exists among the repeated measures for each participant or cluster in longitudinal data. In addition, the GEE approach can accommodate a variety of differential observance patterns for longitudinal data, including scenarios where a participant has only one observation, which allows us to leverage information from all available data. Using the GEE approach, we evaluated odds of falling and falls injury along with the 95% confidence interval, as a function of self-reported hearing status as individuals age. An unstructured correlation structure was used for all models. Effect modification by hearing aid use was determined by including hearing aid use as a covariate and as an interaction with self-reported hearing status in multivariable logistic regression models. Age was treated as a continuous time variable. Missing indicators were included for covariates with missing values under the assumption of missingness at random, wherein the missingness of some self-reported covariates is independent of falls, conditional on the observed survey data (Supplemental table 1 details the baseline demographics and hearing status of the four variables that demonstrated greater than 5% missingness in the data). For logistic regression models, P-values were calculated using a Score statistic. Significance was set to p < 0.05. Sampling weights from HRS were not included, therefore all analyses utilize the unbalanced sample, which is preferable in this cohort for longitudinal models where attrition is present (Kapteyn et al, 2006). Statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC).

RESULTS

Patient Demographics at Time of Entry

There were 17,923 older adults that had at least one observation (i.e., self-report for hearing and falls-related variables) during the seven survey cycles of interest. In Supplemental Table 2, the demographics and health-related variables at the point of entry in the study are listed and described for the study sample as a function of those reporting falls at the time of entry. As seen in Supplemental Table 2, at time of entry, 14.5% (n = 2,591) reported “excellent” hearing, 24.9% (n = 4,471) reported “very good” hearing, 35.9% (n = 6,434) reported “good” hearing, 17.9% (n = 3,214) reported “fair” hearing, and 6.7% (n = 1,194) “poor” hearing. At time of entry, 3.1% (n = 549) reported hearing aid use. Supplemental Table 3 provides a cross-tabulation of hearing aid use by self-reported hearing status. At time of entry, a total of 32.8% (n = 5,872) reported a fall in the past two years. At the time of entry, 10.5% (n = 889) of individuals stated they had experienced a fall that required medical attention in the past two years. Supplemental Tables 4 and 5 further detail the occurrences of falls and falls injury across cycles of the analysis.

Table 1 shows the results of the GEE multivariable logistic regression modeling performed to examine the association between self-reported hearing status and falls. As seen in the table, there was a significant association between self-reported hearing status and falls in both the unadjusted and partially adjusted models (p <0.001 and the fully-adjusted model (p =0.0022). In addition, the models showed an increased odds of falling with increasing self-reported hearing status difficulty. When examining the fully-adjusted model, there was a 20% increased odds of reporting a fall among those respondents reporting “poor” hearing compared to those reporting “excellent” hearing. When we allowed the association to differ by hearing aid use, there was no significant difference in the odds of falling in the fully-adjusted model between hearing aid users and non-users (Table 2, p-interaction = 0.1178), suggesting that hearing aid use did not have a statistically significant impact on the association between self-reported hearing status and falls.

Table 1:

Odds Ratio (OR) for Falls as a Function of Self-Reported Hearing Status Among Study Sample (N= 17,923 individuals* age >65 years) from Health and Retirement Study

Self-Reported Hearing Status Categories OR (95% CI)
Unadjusted Adjusted for covariates1 Adjusted for covariates2
Excellent Hearing (ref) (ref) (ref)
Very Good Hearing 1.08 (1.02, 1.14) 1.06 (1.00, 1.12) 1.06 (1.00, 1.12)
Good Hearing 1.24 (1.17, 1.31) 1.11 (1.05, 1.18) 1.08 (1.02, 1.14)
Fair Hearing 1.51 (1.43, 1.61) 1.22 (1.14, 1.30) 1.12 (1.05, 1.20)
Poor Hearing 2.05 (1.90, 2.20) 1.40 (1.29, 1.51) 1.20 (1.10, 1.30)
p-value <.0001 <.0001 0.0022
Open in a new tab
1

= Adjusted for hearing aid use, age, sex, education, race, ethnicity, eyesight, memory, depression, current smoking status, history of smoking, high blood pressure, history of diabetes, and history of stroke.

2

= Adjusted for hearing aid use, age, sex, education, race, ethnicity, eyesight, memory, depression, current smoking status, history of smoking, high blood pressure, history of diabetes, history of stroke, history of cancer, history of heart disease, self-rated health, self-rated pain, physical inactivity, ADL, IADL, history of arthritis, hospitalization in the past two years, and obese BMI

*

17,923 unique individuals contributed to the total 71,661 observations

Note: Data Source: Health and Retirement Study, Survey cycles 2004-2016

Table 2:

Odds Ratio (OR) for Falls by Self-Reported Hearing Status and Hearing Aid Use in Study Sample (N = 17,923 individuals* age >65 years) from Health and Retirement Study

Self-Reported Hearing Status Categories OR (95% CI)
Unadjusted Adjusted for covariates1 Adjusted for covariates2
Hearing Aid Non-User
Excellent Hearing (ref) (ref) (ref)
Very Good Hearing 1.08 (1.03, 1.15) 1.07 (1.01, 1.13) 1.07 (1.01, 1.13)
Good Hearing 1.24 (1.18, 1.31) 1.13 (1.06, 1.19) 1.09 (1.03, 1.16)
Fair Hearing 1.53 (1.43, 1.62) 1.24 (1.16, 1.32) 1.14 (1.06, 1.22)
Poor Hearing 2.06 (1.91, 2.22) 1.43 (1.32, 1.55) 1.23 (1.13, 1.33)
Hearing Aid User
Excellent Hearing 1.61 (1.18, 2.20) 1.49 (1.08, 2.04) 1.48 (1.07, 2.06)
Very Good Hearing 1.50 (1.26, 1.79) 1.26 (1.04, 1.51) 1.22 (1.00, 1.48)
Good Hearing 1.61 (1.42, 1.83) 1.29 (1.13, 1.47) 1.23 (1.07, 1.40)
Fair Hearing 1.92 (1.65, 2.24) 1.35 (1.15, 1.59) 1.22 (1.03, 1.44)
Poor Hearing 2.75 (2.28, 3.33) 1.56 (1.28, 1.92) 1.33 (1.07, 1.63)
p-interaction 3 .0002 .1100 .1178
Open in a new tab
1

= Adjusted for hearing aid use, age, sex, education, race, ethnicity, eyesight, memory, depression, current smoking status, history of smoking, high blood pressure, history of diabetes, and history of stroke.

2

= Adjusted for hearing aid use, age, sex, education, race, ethnicity, eyesight, memory, depression, current smoking status, history of smoking, high blood pressure, history of diabetes, history of stroke, history of cancer, history of heart disease, self-rated health, self-rated pain, physical inactivity, ADL, IADL, history of arthritis, hospitalization in the past two years, and obese BMI

3

= p-interaction is a test for the interaction between hearing aid use and Self-Reported Hearing Status groups;

*

17,923 unique individuals contributed to the total 71,653 observations for the analysis

Note: Data Source: Health and Retirement Study, Survey cycles 2004-2016

We repeated the aforementioned analyses using self-reported falls-related injury instead of self-reported falls. When examining the association between self-reported hearing status and falls-related injury, a significant association was observed in the unadjusted and partially adjusted regression models, but was no longer significant in the fully-adjusted regression model (Table 3). When the association was allowed to differ by hearing aid use, there was no significant difference in the odds of a falls-related injury between hearing aid users and non-users (Table 4, p = 0.84), suggesting that hearing aid use did not have a statistically significant impact on the association between self-reported hearing status and falls-related injury.

Table 3:

Odds Ratio (OR) for Falls-Related Injury by Self-Reported Hearing Status in Study Sample (N = 17,923 individuals* age >65 years) from Health and Retirement Study

Self-Reported Hearing Status Categories OR (95% CI)
Unadjusted Adjusted for covariates1 Adjusted for covariates2
Excellent Hearing (ref) (ref) (ref)
Very Good Hearing 1.03 (0.94, 1.12) 1.02 (0.93, 1.11) 1.02 (0.94, 1.12)
Good Hearing 1.11 (1.02, 1.21) 1.02 (0.94, 1.12) 0.99 (0.90, 1.08)
Fair Hearing 1.38 (1.27, 1.51) 1.14 (1.04, 1.26) 1.05 (0.95, 1.15)
Poor Hearing 1.94 (1.75, 2.15) 1.29 (1.15, 1.45) 1.10 (0.98, 1.24)
p-value <.0001 <.0001 0.3235
Open in a new tab
1

= Adjusted for hearing aid use, age, sex, education, race, ethnicity, eyesight, memory, depression, current smoking status, history of smoking, high blood pressure, history of diabetes, and history of stroke.

2

= Adjusted for hearing aid use, age, sex, education, race, ethnicity, eyesight, memory, depression, current smoking status, history of smoking, high blood pressure, history of diabetes, history of stroke, history of cancer, history of heart disease, self-rated health, self-rated pain, physical inactivity, ADL, IADL, history of arthritis, hospitalization in the past two years, and obese BMI

*

17,923 unique individuals contributed to the total 71,661 observations

Note Data Source: Health and Retirement Study, Survey cycles 2004-2016;

Table 4:

Odds Ratio (OR) for Falls-Related Injury by Self-Reported Hearing Status and Hearing Aid Use in Study Sample (N = 17,923 individuals* age >65 years) from Health and Retirement Study

Self-Reported Hearing Status Categories OR (95% CI)
Unadjusted Adjusted for covariates1 Adjusted for covariates2
Hearing Aid Non-Users
Excellent Hearing (ref) (ref) (ref)
Very Good Hearing 1.03 (0.95, 1.13) 1.03 (0.94, 1.12) 1.04 (0.95, 1.13)
Good Hearing 1.11 (1.02, 1.21) 1.04 (0.95, 1.13) 1.00 (0.92, 1.10)
Fair Hearing 1.37 (1.25, 1.50) 1.16 (1.05, 1.27) 1.06 (0.96, 1.17)
Poor Hearing 1.93 (1.74, 2.15) 1.34 (1.20, 1.51) 1.15 (1.02, 1.29)
Hearing Aid Users
Excellent Hearing 1.51 (0.97, 2.36) 1.44 (0.91, 2.29) 1.52 (0.92, 2.50)
Very Good Hearing 1.34 (1.01, 1.76) 1.13 (0.85, 1.50) 1.07 (0.79, 1.44)
Good Hearing 1.41 (1.17, 1.71) 1.15 (0.95, 1.39) 1.10 (0.90, 1.34)
Fair Hearing 2.01 (1.64, 2.47) 1.40 (1.13, 1.73) 1.27 (1.01, 1.58)
Poor Hearing 2.58 (2.01, 3.32) 1.33 (1.02, 1.73) 1.15 (0.88, 1.49)
p-interaction 3 0.54 0.88 0.84
Open in a new tab
1

= Adjusted for hearing aid use, age, sex, race, ethnicity, eyesight, memory, depression, current smoking status, history of smoking, high blood pressure, history of diabetes, and history of stroke.

2

= Adjusted for hearing aid use, age, sex, education, race, ethnicity, eyesight, memory, depression, current smoking status, history of smoking, high blood pressure, history of diabetes, history of stroke, history of cancer, history of heart disease, self-rated health, self-rated pain, physical inactivity, ADL, IADL, history of arthritis, hospitalization in the past two years, and obese BMI

3

= p-interaction is a test for the interaction between hearing aid use and Self-Reported Hearing Status groups

*

17,923 unique individuals contributed to the total 71,653 observations

Note: Data Source: Health and Retirement Study, Survey cycles 2004-2016

DISCUSSION

Identifying modifiable risk factors for falls is critically important as falls result in significant injuries, loss of independence, and substantial healthcare burden including increased health care utilization and associated costs. Each year, approximately 3 million older adults are treated in emergency departments for falls and one out of five falls causes a serious injury such as a broken bone (e.g., hip fracture) or head injury (Alexander, Rivara, & Wolf, 1992; Florence et al., 2018a). The total annual medical costs for fatal and non-fatal falls exceed more than $50 billion among older individuals (Florence et al., 2018). Given the recent evidence from Lin and Ferrucci (2012) that demonstrated an association between hearing loss and falls, the current study aimed 1) to determine whether the association between self-rated hearing and falls differed by hearing aid use, and 2) to extend the work to falls-related injury. Our study used GEE which provides robust unbiased estimates on the longitudinal outcomes of falls and falls-injury. These models incorporate the correlations between repeated outcomes of the individual. Our results do not support our original hypothesis that the association between hearing loss and falls would be reduced in hearing aid users relative to non-hearing aid users. Our results do, however, confirm prior epidemiological studies showing the independent association between hearing loss and falls (Gopinath et al., 2016; Jiam et al., 2016; Lin & Ferrucci, 2012; Lopez et al., 2011) and expand upon prior studies by demonstrating that self-reported hearing status, as opposed to hearing assessed by pure-tone audiometry, is also associated with increased odds of falling (Gopinath et al 2016, Jiam et al 2016). Interestingly, our results showed no significant association between self-related hearing status and history of falls-related injury. In our partially adjusted model (i.e., adjusted for demographics, eyesight, memory, depression, smoking status, diabetes and stroke), the association was significant; however, did not remain significant when we adjust for additional confounders such as major health concerns, hospitalizations, and ability to perform ADLs/IADLs, and physical activity.

The novel contribution of the current work is that when examining self-reported hearing status and falls, the association between hearing status and falls or falls-related injury does not differ between hearing aid users and nonusers. A study by Kamil and colleagues (2016) examined the impact of hearing aid use on falls. Similar to the findings in our study, their study reported no differences in the odds of falling between hearing aid users and nonusers. They measured pure-tone thresholds in their study and restricted hearing loss to moderate or greater in their analysis. While we did not have audiometric thresholds available in the HRS data, we examined odds of falls and falls-related injury across the full range of self-reported hearing status, from excellent to very poor. Despite the differences between their study and ours, we reached similar conclusions in that the odds of falling did not differ by hearing aid use.

In contrast, Gopinath and colleagues (2016) reported an increased odds of falling for hearing aid users with severe self-perceived hearing handicap. It is important to note that this finding was demonstrated in those who also had concomitant visual impairment, which they defined as best corrected visual acuity in the better eye less than (i.e., poorer) 20/40. Thus, the increased odds of falling found in the Gopinath cohort may be due to the fact that many may have had an increased risk of falls at baseline due to multifactorial causes (e.g., dual-sensory impairment) and/or may have had an underlying visual dysfunction above and beyond age-related changes in acuity. In our model, we controlled for a range of self-reported vision from excellent to very poor and found no evidence of a difference between hearing aid users and nonusers on falls or falls-related injury. Nonetheless, research suggests that visual acuity and visual function are important contributors for balance and falls risk (Dhital, Pey, & Stanford, 2010; Lamoreux et al., 2008; Lord, 2006; Reed-Jones et al., 2013), and thus it was important to consider in our study.

There are several possible reasons that may explain the mixed results regarding the impact of hearing aid use on falls and falls-related injury to date. First, the simple presence or absence of hearing aid use is likely an overly simplistic way to determine whether hearing aid use influences the risk for falls or falls-related injury in patients with hearing loss and/or self-reported hearing status difficulty. Instead, the results may highlight a need to examine hearing aid benefit. While patients reported their self-rated hearing in the aided condition (if a hearing aid user), the question was not specifically designed to assess benefit. Our study may highlight the need for examination of global hearing aid benefit and specific domains of hearing aid benefit and/or the presence of inadequate hearing aid benefit. For example, hearing aid users with severe self-perceived hearing handicap were at greatest risk for falls in the Gopinath study, suggesting that hearing aid use in that sample may not have been optimized as they continued to demonstrate significant residual handicap (i.e., still had significant handicap despite the presence of a hearing aid). Second, the poor visual acuity in their sample may have also exacerbated their falls risk and among dual sensory impairment, optimized hearing treatment may be more important.

These findings coupled with our results and others may suggest the importance of examining the degree of treatment benefit across important functional hearing domains, such as sound localization and listening, rather than simply knowing the presence/absence of hearing aid use. It may be that those individuals with hearing aids who underuse them and/or that do not provide adequate benefit in these important functional hearing domains are the individuals at higher risk of falls. To our knowledge, the influence of hearing aid use and benefit on the odds of falling has not been evaluated. Moreover, this work may also provide indirect evidence that the association between hearing loss and falls may be mediated by a non-auditory factor such as concomitant vestibular loss or some other common cause etiology (e.g., vascular disease). In the case of a non-auditory mediated factor, one would expect that treatment of hearing loss through hearing aids would not significantly impact odds of falling. Current work in our lab continues to research these possibilities.

There are study limitations. First, our study defined hearing aid use by the question “Do you ever wear a hearing aid?” This variable does not allow us to understand wear time or the extent of benefit that the individual received across a variety of outcome domains. It is possible that without this knowledge on use and benefit, the lack of information may mask the magnitude of the effect of hearing aids on odds of falling. While this HRS variable has limitations, it has been used by several studies to study hearing aid use in older adults (Klyn, Mohammed Shaikh, & Dhar, 2020; Maharani, Dawes, Nazroo, Tampubolon, & Pendleton, 2018; M. McKee & Choi, 2017; M. M. McKee et al., 2019) and allows for hypothesis generation for future work. Second, our study sample with hearing aid use was small (n = 549 at study entry). It could be that we are underpowered to detect the true impact of hearing aid use on falls. Third, our study utilized self-report data to characterize the variables of interest, all of which can impose recall bias. Despite the potential for bias, researchers have demonstrated that self-report measures are reliable indicators of hearing impairment (Chou, Dana, Bougatsos, Fleming, & Beil, 2011; Clark, Sowers, Wallace, & Anderson, 1991). Moreover, in the current study we were unable to control for the potential for polypharmacy, high-risk medication use, orthopedic changes, dizziness/vestibular disease, and/or fear of falling, all of which are variables that my influence falls or falls-related injury. Finally, in any longitudinal analysis, there are concerns for missing data and/or attrition. In our study, we designed our analysis to account for the primary sources of attrition in elderly cohort studies (which are age and cognitive impairment). Additionally, based on prior work examining attrition in the HRS study (Kapteyn et al 2006), we used the unbalanced sample rather than the sampling weights. This analytical approach was based on the recommendation of prior literature (Kapteyn et al 2006) who determined that longitudinal analyses using HRS should use the unbalanced sample, rather than incorporating sampling weights, given the rates of attrition differ across demographics. As such, our analyses used the unbalanced samples.

CONCLUSION

To conclude, the unique results of the current study show that when examining self-reported hearing status in a longitudinal sample ranging over 7 biennial cycles spanning 14 years, hearing aid use does not appear to modify the association between self-reported hearing status and falls or falls-related injury. Additional research is needed to examine various hearing aid use and benefit domains to further understand the impacts, if any, that hearing aid use and benefit plays in the independent association between hearing loss and falls, and to examine non-auditory factors that may contribute to the association between hearing loss and falls.

Supplementary Material

Supplemental Material

Supplemental Table 1: Hearing and Demographic Evaluation of HRS Variables (Hearing Aid Use, Self-Reported Memory, Self-Reported Depression, Self-Reported Arthritis) with greater than 5% Missing at Baseline

Supplemental Table 2: Sample Characteristics of Health and Retirement Study (HRS) Participants at the Time of Entry by Falls Status

Supplemental Table 3: Cross-tablulation of self-reported hearing with hearing aid user status for all observations included in Health and Retirement Study (2004-2016) longitudinal models for falls status

Supplemental Table 4: Falls Status across Health and Retirement Study Cycles (2004-2016)

Supplemental Table 5: Fall-related injuries across Health and Retirement Study Cycles 2004-2016

Acknowledgements:

K.M.R. oversaw the design, data acquisition, data analysis/interpretation and drafted the manuscript. S.B.P. contributed to the design of the project, data acquisition, data analysis and interpretation and drafting and revising the manuscript critically for important intellectual content. M.K. contributed to the analysis and interpretation of the data and revising the manuscript critically for important intellectual content. A.G. contributed to the data acquisition, analysis, and interpretation of the revised manuscript, as well as revising the manuscript critically for important intellectual content. J.P. contributed to the design of the work, interpretation of the data, and revising the manuscript critically for important intellectual content, S.E.K. contributed to the analysis and interpretation of the data and revising the manuscript critically for important intellectual content, J.S.W. contributed to the design of the project, data acquisition, data interpretation and drafting and revising the manuscript critically for important intellectual content. S.L.S. contributed to the design of the project, data interpretation and drafting and revising the manuscript critically for important intellectual content. All authors provided final approval of the version to be published.

Conflicts of Interest and Source of Funding

The Duke Biostatics, Epidemiology, and Research Design (BERD) Methods Core’s support of this project was made possible (in part) by Grant Number UL1TR002553 from the National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health (NIH), and NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NCATS or NIH.

REFERENCES

  1. Alexander BH, Rivara FP, & Wolf ME (1992). The cost and frequency of hospitalization for fall-related injuries in older adults. American Journal of Public Health, 82(7), 1020–1023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Berge JE, Nordahl SHG, Aarstad HJ, & Goplen FK (2019). Hearing as an Independent Predictor of Postural Balance in 1075 Patients Evaluated for Dizziness. Otolaryngology–Head and Neck Surgery, 0194599819844961. [DOI] [PubMed] [Google Scholar]
  3. Chatfield MD, Brayne CE, Matthews FE, 2005. A systematic literature review of attrition between waves in longitudinal studies in the elderly shows a consistent pattern of dropout between differing studies. J. Clin. Epidemiol 58, 13–19. [DOI] [PubMed] [Google Scholar]
  4. Chou R, Dana T, Bougatsos C, Fleming C, & Beil T (2011). Screening for hearing loss in adults ages 50 years and older: a review of the evidence for the US Preventive Services Task Force. [PubMed]
  5. Clark K, Sowers M, Wallace RB, & Anderson C (1991). The accuracy of self-reported hearing loss in women aged 60–85 years. American journal of epidemiology, 134(7), 704–708. [DOI] [PubMed] [Google Scholar]
  6. Davis JC, Bryan S, Best JR, Li LC, Hsu CL, Gomez C, … Liu-Ambrose T. (2015). Mobility predicts change in older adults’ health-related quality of life: evidence from a Vancouver falls prevention prospective cohort study. Health and quality of life outcomes, 13(1), 101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Deandrea S, Lucenteforte E, Bravi F, Foschi R, La Vecchia C, & Negri E (2010). Risk factors for falls in community-dwelling older people:” a systematic review and meta-analysis”. Epidemiology, 658–668. [DOI] [PubMed] [Google Scholar]
  8. Delbaere K, Crombez G, Vanderstraeten G, Willems T, & Cambier D (2004). Fear-related avoidance of activities, falls and physical frailty. A prospective community-based cohort study. Age and ageing, 33(4), 368–373. [DOI] [PubMed] [Google Scholar]
  9. Deviterne D, Gauchard GC, Jamet M, Vançon G, & Perrin PP (2005). Added cognitive load through rotary auditory stimulation can improve the quality of postural control in the elderly. Brain research bulletin, 64(6), 487–492. [DOI] [PubMed] [Google Scholar]
  10. Dhalwani NN, Fahami R, Sathanapally H, Seidu S, Davies MJ, & Khunti K (2017). Association between polypharmacy and falls in older adults: a longitudinal study from England. BMJ open, 7(10), e016358. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dhital A, Pey T, & Stanford MR (2010). Visual loss and falls: a review. Eye, 24(9), 1437. [DOI] [PubMed] [Google Scholar]
  12. Dozza M, Horak FB, & Chiari L (2007). Auditory biofeedback substitutes for loss of sensory information in maintaining stance. Experimental Brain Research, 178(1), 37–48. [DOI] [PubMed] [Google Scholar]
  13. Dunn JE, Furner SE, & Miles TP (1993). Do falls predict institutionalization in older persons? An analysis of data from the Longitudinal Study of Aging. Journal of aging and health, 5(2), 194–207. [DOI] [PubMed] [Google Scholar]
  14. Ernst A, Basta D, Mittmann P, & Seidl RO (2020). Can hearing amplification improve presbyvestibulopathy and/or the risk-to-fall?. European archives of oto-rhino-laryngology, 1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Florence CS, Bergen G, Atherly A, Burns E, Stevens J, & Drake C (2018). Medical costs of fatal and nonfatal falls in older adults. Journal of the American Geriatrics Society, 66(4), 693–698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gopinath B, McMahon CM, Burlutsky G, & Mitchell P (2016). Hearing and vision impairment and the 5-year incidence of falls in older adults. Age Ageing, 45(3), 409–414. doi: 10.1093/ageing/afw022 [DOI] [PubMed] [Google Scholar]
  17. Guirguis-Blake JM, Michael YL, Perdue LA, Coppola EL, & Beil TL (2018). Interventions to prevent falls in older adults: updated evidence report and systematic review for the US Preventive Services Task Force. Jama, 319(16), 1705–1716. [DOI] [PubMed] [Google Scholar]
  18. Haran MJ, Cameron ID, Ivers RQ, Simpson JM, Lee BB, Tanzer M, … Lord SR (2010). Effect on falls of providing single lens distance vision glasses to multifocal glasses wearers: VISIBLE randomised controlled trial. Bmj, 340, c2265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hartholt KA, van Beeck EF, Polinder S, van der Velde N, van Lieshout EM, Panneman MJ, … Patka P (2011). Societal consequences of falls in the older population: injuries, healthcare costs, and long-term reduced quality of life. Journal of Trauma and Acute Care Surgery, 71(3), 748–753. [DOI] [PubMed] [Google Scholar]
  20. Ivers R, Cumming R, & Mitchell P (2002). Poor vision and risk of falls and fractures in older Australians: the Blue Mountains Eye Study. New South Wales public health bulletin, 13(2), 8–10. [DOI] [PubMed] [Google Scholar]
  21. Jiam NTL, Li C, & Agrawal Y (2016). Hearing loss and falls: A systematic review and meta-analysis. The Laryngoscope, 126(11), 2587–2596. [DOI] [PubMed] [Google Scholar]
  22. Kamil RJ, Betz J, Powers BB, Pratt S, Kritchevsky S, Ayonayon HN, … Martin K (2016). Association of hearing impairment with incident frailty and falls in older adults. Journal of aging and health, 28(4), 644–660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kapteyn Arie and Michaud Pierre-Carl and Smith James P. and van Soest Arthur H. O., Effects of Attrition and Non-Response in the Health and Retirement Study (August 2006). IZA Discussion Paper No. 2246, RANSD Working Paper No. WR-407, Available at SSRN: https://ssrn.com/abstract=928369 [Google Scholar]
  24. Klyn NAM, Mohammed Shaikh Z, & Dhar S (2020). Health Literacy and Self-Reported Hearing Aid Use in the Health and Retirement Study. Ear Hear, 41(2), 386–394. doi: 10.1097/aud.0000000000000770 [DOI] [PubMed] [Google Scholar]
  25. Kulmala J, Viljanen A, Sipilä S, Pajala S, Pärssinen O, Kauppinen M, … Rantanen T (2009). Poor vision accompanied with other sensory impairments as a predictor of falls in older women. Age and ageing, 38(2), 162–167. [DOI] [PubMed] [Google Scholar]
  26. Lacerda CF, Silva LO, de Tavares Canto RS, & Cheik NC (2012). Effects of hearing aids in the balance, quality of life and fear to fall in elderly people with sensorineural hearing loss. Int Arch Otorhinolaryngol, 16(2), 156–162. doi: 10.7162/s1809-97772012000200002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lamoreux EL, Chong E, Wang JJ, Saw SM, Aung T, Mitchell P, & Wong TY (2008). Visual impairment, causes of vision loss, and falls: the Singapore Malay Eye Study. Investigative ophthalmology & visual science, 49(2), 528–533. [DOI] [PubMed] [Google Scholar]
  28. Li F, Fisher KJ, Harmer P, McAuley E, & Wilson NL (2003). Fear of falling in elderly persons: association with falls, functional ability, and quality of life. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 58(5), P283–P290. [DOI] [PubMed] [Google Scholar]
  29. Li KZ, & Lindenberger U (2002). Relations between aging sensory/sensorimotor and cognitive functions. Neuroscience & Biobehavioral Reviews, 26(7), 777–783. [DOI] [PubMed] [Google Scholar]
  30. Lin FR, & Ferrucci L (2012). Hearing loss and falls among older adults in the United States. Archives of internal medicine, 172(4), 369–371. Retrieved from https://jamanetwork.com/journals/jamainternalmedicine/articlepdf/1108740/ilt120001_369_371.pdf [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lopez D, McCaul KA, Hankey GJ, Norman PE, Almeida OP, Dobson AJ, … Flicker L (2011). Falls, injuries from falls, health related quality of life and mortality in older adults with vision and hearing impairment—is there a gender difference? Maturitas, 69(4), 359–364. [DOI] [PubMed] [Google Scholar]
  32. Lord SR (2006). Visual risk factors for falls in older people. Age and ageing, 35(suppl_2), ii42–ii45. [DOI] [PubMed] [Google Scholar]
  33. Maharani A, Dawes P, Nazroo J, Tampubolon G, & Pendleton N (2018). Longitudinal Relationship Between Hearing Aid Use and Cognitive Function in Older Americans. Journal of the American Geriatrics Society, 66(6), 1130–1136. doi: 10.1111/jgs.15363 [DOI] [PubMed] [Google Scholar]
  34. Mahmoudi E, Basu T, Langa K, McKee MM, Zazove P, Alexander N, & Kamdar N (2019). Can Hearing Aids Delay Time to Diagnosis of Dementia, Depression, or Falls in Older Adults? J Am Geriatr Soc. doi: 10.1111/jgs.16109 [DOI] [PubMed] [Google Scholar]
  35. McKee M, & Choi H (2017). Hearing health equity: Social determinants of hearing aid receipts. Paper presented at the MedEq, Detroit, MI. [Google Scholar]
  36. McKee MM, Choi H, Wilson S, DeJonckheere MJ, Zazove P, & Levy HG (2019). Determinants of Hearing Aid Use Among Older Americans With Hearing Loss. Gerontologist, 59(6), 1171–1181. doi: 10.1093/geront/gny051 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mergner T, Maurer C, & Peterka R (2003). A multisensory posture control model of human upright stance. Progress in brain research, 142, 189–202. [DOI] [PubMed] [Google Scholar]
  38. Montero-Odasso M, Sarquis-Adamson Y, Song HY, Bray NW, Pieruccini-Faria F, & Speechley M (2019). Polypharmacy, Gait Performance, and Falls in Community-Dwelling Older Adults. Results from the Gait and Brain Study. Journal of the American Geriatrics Society, 67(6), 1182–1188. [DOI] [PubMed] [Google Scholar]
  39. Negahban H, Bavarsad Cheshmeh ali M, & Nassadj G (2017). Effect of hearing aids on static balance function in elderly with hearing loss. Gait & Posture, 58, 126–129. doi: 10.1016/j.gaitpost.2017.07.112 [DOI] [PubMed] [Google Scholar]
  40. Peel NM (2011). Epidemiology of falls in older age. Canadian Journal on Aging/La Revue canadienne du vieillissement, 30(1), 7–19. [DOI] [PubMed] [Google Scholar]
  41. Phillips NA (2016). The implications of cognitive aging for listening and the framework for understanding effortful listening (FUEL). Ear and Hearing, 37, 44S–51S. [DOI] [PubMed] [Google Scholar]
  42. Reed-Jones RJ, Solis GR, Lawson KA, Loya AM, Cude-Islas D, & Berger CS (2013). Vision and falls: a multidisciplinary review of the contributions of visual impairment to falls among older adults. Maturitas, 75(1), 22–28. [DOI] [PubMed] [Google Scholar]
  43. Rubenstein LZ (2006). Falls in older people: epidemiology, risk factors and strategies for prevention. Age and ageing, 35(suppl_2), ii37–ii41. [DOI] [PubMed] [Google Scholar]
  44. Rumalla K, Karim AM, & Hullar TE (2015). The effect of hearing aids on postural stability. The Laryngoscope, 125(3), 720–723. [DOI] [PubMed] [Google Scholar]
  45. Servais M (2010). Overview of HRS Public Data Files for Cross-sectional and Longitudinal Analysis. Retrieved from https://hrs.isr.umich.edu/documentation/survey-design
  46. Shayman CS, Earhart GM, & Hullar TE (2017). Improvements in gait with hearing aids and cochlear implants. Otology & Neurotology, 38(4), 484–486. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Shayman CS, Mancini M, Weaver TS, King LA, & Hullar TE (2017). The contribution of cochlear implants to postural stability. The Laryngoscope. [DOI] [PubMed] [Google Scholar]
  48. Staff H (2008). HRS Design History. Retrieved from https://hrs.isr.umich.edu/documentation/survey-design
  49. Staff H (2017). Sample Size and Response Rate. Retrieved from https://hrs.isr.umich.edu/documentation/survey-design
  50. Stenhagen M, Ekström H, Nordell E, & Elmståhl S (2014). Accidental falls, health-related quality of life and life satisfaction: a prospective study of the general elderly population. Archives of gerontology and geriatrics, 58(1), 95–100. [DOI] [PubMed] [Google Scholar]
  51. Stevens MN, Barbour DL, Gronski MP, & Hullar TE (2016). Auditory contributions to maintaining balance. Journal of Vestibular Research, 26(5-6), 433–438. [DOI] [PubMed] [Google Scholar]
  52. Tinetti ME, & Williams CS (1997). Falls, injuries due to falls, and the risk of admission to a nursing home. New England journal of medicine, 337(18), 1279–1284. [DOI] [PubMed] [Google Scholar]
  53. Vellas B, Cayla F, Bocquet H, De Pemille F, & Albarede J (1987). Prospective study of restriction of acitivty in old people after falls. Age and ageing, 16(3), 189–193. [DOI] [PubMed] [Google Scholar]
  54. Vitkovic J, Le C, Lee SL, & Clark RA (2016). The Contribution of Hearing and Hearing Loss to Balance Control. Audiol Neurootol, 21(4), 195–202. doi: 10.1159/000445100 [DOI] [PubMed] [Google Scholar]
  55. Vu T, Finch CF, & Day L (2011). Patterns of comorbidity in community-dwelling older people hospitalised for fall-related injury: a cluster analysis. BMC geriatrics, 11(1), 45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Weaver TS, Shayman CS, & Hullar TE (2017). The effect of hearing aids and cochlear implants on balance during gait. Otology & Neurotology, 38(9), 1327–1332. [DOI] [PubMed] [Google Scholar]
  57. Zuniga MG, Dinkes RE, Davalos-Bichara M, Carey JP, Schubert MC, King WM, … & Agrawal Y (2012). Association between hearing loss and saccular dysfunction in older individuals. Otology & Neurotology, 33(9), 1586. [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

Supplemental Material

Supplemental Table 1: Hearing and Demographic Evaluation of HRS Variables (Hearing Aid Use, Self-Reported Memory, Self-Reported Depression, Self-Reported Arthritis) with greater than 5% Missing at Baseline

Supplemental Table 2: Sample Characteristics of Health and Retirement Study (HRS) Participants at the Time of Entry by Falls Status

Supplemental Table 3: Cross-tablulation of self-reported hearing with hearing aid user status for all observations included in Health and Retirement Study (2004-2016) longitudinal models for falls status

Supplemental Table 4: Falls Status across Health and Retirement Study Cycles (2004-2016)

Supplemental Table 5: Fall-related injuries across Health and Retirement Study Cycles 2004-2016

NIHMS1838172-supplement-Supplemental_Material.docx (29KB, docx)

RESOURCES