Self‐Reported Hearing Status Is Associated with Lower Limb Physical Performance, Perceived Mobility, and Activities of Daily Living in Older Community‐Dwelling Men and Women

To explore the associations between self‐reported hearing problems and physical performance and self‐reported difficulties in mobility and activities of daily living (ADLs) in community‐dwelling older adults.

H earing impairment is a major health concern in older adults, affecting more than 50% of adults aged 70 and older. 1,2 In addition to complicating interaction with other people, poor hearing may have consequences that extend beyond communication. Impaired hearing may make walking more uncertain, because acoustic cues assist in perception of the environment while moving. 3 Persons with hearing impairment may also reduce their participation in social activities, 4,5 which in turn may negatively affect their physical performance by reducing overall physical activity.
Good physical performance is essential for older adults' independent functioning [6][7][8] and safe moving 9,10 in everyday life. The associations between hearing and physical performance have been investigated in only a few studies. One study found no association between self-reported hearing loss and Short Physical Performance Battery (SPPB) scores in American Indians aged 55 and older. 11 Two studies in older adults have shown that persons with audiometrically determined hearing impairment have a slower walking speed than those with normal hearing. 12,13 The participants in these studies were all younger than 77. Furthermore, self-reported mobility, activity of daily living (ADL), and instrumental ADL (IADL) difficulties have been shown to be greater in older persons with hearing problems in some studies but not all. [13][14][15][16][17][18][19] Furthermore, studies that have specifically analyzed adults aged 75 and older have not found any association between hearing and ADLs. 17,19 Given the growing prevalence of hearing problems with increasing age, it is vitally important to study the correlates of hearing difficulties in older adults. The primary purpose of the present study was to explore the associations between self-reported hearing problems and physical performance in community-dwelling older adults. The association between self-reported hearing problems and perceived mobility and ADL difficulties was also studied.

Design and Sample
The analyses made use of cross-sectional data gathered for the Life-Space Mobility in Old Age (LISPE) project, a study of community-dwelling older adults that has been described in detail elsewhere. 20 Briefly, a sample of 2,550 individuals aged 75 to 90 living in the municipalities of Muurame and Jyv€ askyl€ a in central Finland was drawn from the population register. A telephone interview was used to screen eligible participants based on the inclusion criteria: living independently, residing in the recruitment area, being able to communicate, and being willing to participate in the study. The final data set comprised 848 individuals who participated in structured interviews and objective assessments of physical performance in their homes. The ethical committee of the University of Jyv€ askyl€ a approved the LISPE project. Subjects provided informed consent at the start of the home interview.

Hearing
Hearing was assessed according to answers to the question: "Do you have difficulty hearing when conversing with another person in a noisy environment?" 4 Response categories were no difficulty; sometimes, some difficulty; and yes, major difficulty. Participants were asked to estimate their level of difficulty when using a hearing aid if they had one.

Physical Performance
Lower limb physical performance was assessed using the SPPB, 21,22 which comprises three tests that assess standing balance, walking speed over 2.44 meters (m/s), and timed chair stands (five times, seconds). Each test was rated from 0 to 4 points. Established age-and sex-specific cutoff points were used in rating walking speed and chair stands tests. 21 A sum score was calculated (range 0-12, higher scores indicating better physical performance) when at least two tests were completed. A SPPB total score was missing in nine participants, balance in 30 participants, walking speed in 24 participants, and chair stand time in 48 participants. The absolute values for walking speed and chair stands time, instead of subscores, were used in the analyses of the individual SPPB tests.

Perceived Mobility
Participants were asked to rate the level of difficulty of four mobility tasks: moving indoors, stair-climbing, 0.5km walk, and 2-km walk. Five response categories were used, ranging from able without difficulty to unable even with the help of another person. 20

Functional Ability
Functional status was assessed using a 14-item self-report questionnaire for ADLs 23 comprising five ADL and nine IADL tasks. Participants were asked to rate their ability to perform each task on a five-point scale ranging from able without difficulty (zero points) to unable even with help of another person (four points). ADL and IADL scores were calculated by summing the scores for the individual ADL and IADL tasks.

Potential Confounders
Age and sex were obtained from the population register. Cognitive function was assessed using the Mini-Mental State Examination (MMSE). 24 Participants were asked whether they had sufficient financial resources to meet their needs and how many years of education they had completed. 20 Body mass index (BMI, kg/m 2 ) was calculated based on self-reported height and weight. Self-reported diseases were obtained from a list of 22 physician-diagnosed chronic diseases and an open question. 25 Diseases that could theoretically be linked to hearing problems and physical performance (diabetes mellitus; cancer; locomotor, rheumatic, cardiac, circulatory, neurological diseases) were chosen as potential covariates.

Data Analysis
Associations between hearing, physical performance, and ADL and IADL sum scores were analyzed using generalized linear models, using the gamma log-link option for the response variable owing to the nonnormal distributions of the response variables. Because this analysis cannot handle response variables with zero values, one was added to the SPPB score and the ADL and IADL sum scores. The associations between hearing problems and the SPPB balance subscore and self-reported mobility were studied using ordinal logistic regression analysis. Moving indoors was an exception; it had to be dichotomized because of the small number of observations in the categories indicating greatest disability, and consequently, binary logistic regression analysis was used. In the generalized linear models and ordinal regression analysis, P-values are given for comparisons between the group with good hearing and all other groups. In all models, the inclusion criterion for the possible confounders was an association (P ≤ .20) with the predictor and the response variable. Of the potential confounders, age; years of education; cognitive functioning; and cardiac, circulatory, and locomotor diseases met this criterion for all response variables and were used as covariates in all analyses. The data were analyzed using IBM SPSS Statistics for Windows version 20.0 (IBM Corp., Armonk, NY). Statistical significance was set at P < .05.

RESULTS
Background characteristics of the participants are given in Table 1. When examining the associations between hearing problems and SPPB total score, the crude models showed that persons reporting major hearing problems had a significantly lower SPPB total score than those reporting good hearing (estimated marginal mean 9.5 vs 10.9, P < .001).
Persons reporting some hearing problems did not differ in SPPB total score from those reporting good hearing (10.6 vs 10.9, P = .17). Adjustment (age; years of education; cognitive functioning; cardiac, circulatory, locomotor diseases) did not materially change the differences between persons with major (9.8 vs 10.9, P = .009) or some (10.7 vs 10.9, P = .46) hearing problems and those with good hearing (Figure 1). Hearing problems were also associated with the SPPB subtests. In the crude and adjusted models, persons reporting major hearing problems had significantly slower walking speed than those reporting good hearing (0.75 vs 0.90 m/s, P < .001 and 0.80 vs 0.88 m/s, P = .008, respectively) (Figure 1). Also, in the crude and adjusted models, persons reporting major hearing problems had significantly longer chair stands time than those reporting good hearing (16.7 vs 13.1 seconds, P < .001 and 15.9 vs 13.2 seconds, P < .001, respectively). In the crude model, persons with some hearing problems had significantly longer chair stands time (13.7 vs 13.1 seconds, P = .04) and less likelihood of a higher balance score (OR=0.44, P = .001) than those reporting good hearing, but these differences became non-significant after adjustment (13.7 vs 13.2 seconds, P = .15 for chair stands time and OR=0.69, P = .15 for balance score).

Perceived Mobility
In the crude models, persons with major hearing problems had statistically significantly greater odds for more difficulties in stair-climbing (OR = 3.9), 0.5-km walk (OR = 2.8), and 2-km walk (OR = 3.3) but not moving indoors than those reporting good hearing (Table 2). After adjustment, the differences for stair-climbing (OR = 2.8) and the 2-km walk (OR = 2.1) remained statistically significant, but the difference for the 0.5-km walk was no longer significant (OR = 1.7). In the adjusted models, persons with some hearing problems did not significantly differ from those with good hearing on any of the perceived mobility variables.

ADLs and IADLs
The crude models showed that persons who reported major hearing problems had significantly more ADL and IADL difficulties than those who reported good hearing (estimated marginal mean 2.0 vs 1.4, P < .001 and 7.1 vs 3.8, P < .001, respectively). Also, persons who reported some hearing problems had significantly more IADL (4.4 vs 3.8, P = .04) but not ADL difficulties than persons reporting good hearing. After adjustment for the covariates, the difference in ADL (1.8 vs 1.4, P = .002) and IADL (4.6 vs 3.4, P = .002) scores between persons with major hearing problems and those with good hearing remained statistically significant.

DISCUSSION
This study showed that self-reported hearing problems are associated with physical performance in community-dwelling older adults aged 75 to 90. Persons who reported major hearing problems had more difficulties in mobility, ADLs, and IADLs than those who reported good hearing. Persons who reported only some hearing problems did not differ from those who reported good hearing in any of the variables of physical performance, perceived mobility, or ADLs after controlling for the effects of potential confounders.   and 95% CI for higher score in balance test in groups, some problems in hearing and major problems in hearing compared with good hearing (OR = 1). In SPPB total and balance score, higher scores indicate better performance. One is added to the SPPB total score owing to the data analysis method. The analyses are adjusted for age; years of education; cognitive functioning; and cardiac, circulatory, and locomotor diseases. To the knowledge of the authors of the current study, only one previous study has investigated the associations between hearing and the SPPB. 11 It found no significant association between self-reported hearing loss and SPPB total score in 328 community-dwelling American Indians aged 55 and older. The purpose of that study was to examine a variety of health conditions-not just hearing loss-and it appears that the sample size in their study was too small to detect differences between hearing groups, although a trend could be seen. The difference between persons with major hearing problems and good hearing on the SPPB total score in the present study can be considered clinically meaningful in older adults. 26 The associations between self-reported hearing problems and the individual tests constituting the SPPB (balance, walking speed, chair stands) were also explored. With regard to walking speed, the results were consistent with those previously reported in younger cohorts. In one study, women with hearing impairment had 0.1-m/s slower walking speed than women with normal hearing, 13 whereas in another study, a 25-dB greater pure-tone average was associated with 0.05-m/s slower walking speed. 12 The current study finding suggests that the relationship between hearing and walking speed, observed in younger old persons, persists after age 75. A new finding of the present study was that persons with major hearing problems had poorer performance on the chair stands test than persons with good hearing, although no significant differences were found in the balance test scores between the hearing groups after adjusting the analyses for relevant confounders. This was unexpected, because hearing and balance can theoretically be linked through proximity of the anatomical structures in the inner ear and, moreover, because pure-tone average and postural sway were correlated in a previous study. 27 Previous studies that have shown that self-reported 15 and audiometrically determined 13 hearing impairment is associated with difficulties in mobility tasks support the current results regarding perceived mobility, although this has not been confirmed in all studies. 16 The current results for different mobility tasks suggest that hearing impairment may be associated with difficulties in more physically demanding tasks. The results support previous findings in younger old cohorts showing that hearing is associated with ADLs and IADLs, although previous studies specifically investigating older cohorts (≥75) have not found significant relationships between hearing and ADLs and IADLs. It may be that the association between hearing and ADLs and IADLs is not particularly strong and may be overridden by that of other health conditions, which could explain the variation observed between study populations. Although minor hearing problems were not independently associated with any of the measures of physical functioning in this cross-sectional analysis, they may nevertheless predict emerging physical disability.
There are several possible explanations for the relationship between hearing and physical performance and mobility. First, pathophysiological processes in the vasculature may negatively influence hearing and lower limb performance. Second, communication difficulties caused by poor hearing reduce participation in different activities, 4,5 and reduced general physical activity due to reduced participation might in turn contribute to decline in physical performance. Third, impaired hearing might make moving more uncertain by disturbing perception of the environment. 3 Fourth, according to the analyses, cognitive function played a role in the relationship between hearing and physical performance. Previous studies have suggested that hearing problems may lead to cognitive decline, 28 which in turn, has been found to be associated with mobility decline. 29 Fifth, in persons with poor hearing, more cognitive resources may be engaged in processing auditory information, thereby reducing the resources available for moving. 30 This study has several strengths. The sample was population based, and the interviews were conducted in participants' homes, facilitating the participation of frailer older people, although persons with mobility problems were more likely to decline participation. 20 Moreover, to be included in the study, participants were required to be able to communicate during the home interview. This criterion probably decreased the number of persons with severe hearing impairment. In addition, the cross-sectional design limits any causal relationships that may be drawn between the variables investigated. It was not possible to assess hearing using an audiometer, but self-reports were used. However, it is not known which hearing parameter, perceived or audiometrically assessed, is the more relevant determinant of functioning in older adults.
In conclusion, older adults with perceived major hearing problem have poorer lower limb performance, mobility, and ADL function than those who perceive their hearing as good. Further studies are needed to determine whether poor hearing is a risk factor for decline in physical performance and whether audiologic rehabilitation has positive effects on older persons' physical performance and functioning.