Combined effect of self-perceived voice disorders and hearing loss on social acts in older adults: a 1-year longitudinal study

Abstract

[Purpose] Self-perceived voice disorders (VD) and hearing loss (HL) are common communication barriers in older adults, yet their combined impact on social participation remains underexplored. This study examined the additive effects of self-perceived VD and HL on frequency of going out (FGO) and social interactions in community-dwelling older Japanese adults. [Participants and Methods] In this longitudinal study, 134 adults aged ≥65 years were surveyed at baseline and one year later. Self-perceived VD and HL were assessed using Voice Handicap Index-10 (cutoff ≥5) and Hearing Handicap Inventory for the Elderly-Screening Version (cutoff ≥8). FGO was measured as days per week participants went outside, and social interaction was assessed using the Lubben Social Network Scale-6. [Results] Repeated-measures analyses showed that self-perceived VD was associated with decreased FGO. Participants with both self-perceived VD and HL showed an additional decrease of about 1 day per week after one year. Participants with either impairment alone or with neither showed no significant change. No significant associations were observed between VD or HL and LSNS-6 family or friend subscale scores. [Conclusion] Co-occurring self-perceived VD and HL may exacerbate social isolation by limiting expressive and receptive communication. Addressing even one of these conditions could help maintain mobility and social interaction.

INTRODUCTION

The global population is rapidly ageing, resulting in an increasing number of older adults who experience age-related communication challenges¹⁾. Among these challenges, voice disorders (VD), commonly referred to as presbyphonia, have garnered increasing attention due to their prevalence and impact on daily life²⁾. Physiological changes, such as muscle atrophy of the vocal folds, alterations in the viscoelastic properties of the vocal fold mucosa, reduced respiratory support, and hormonal shifts, can contribute to incomplete glottal closure, hoarseness, reduced vocal intensity, and other vocal impairments^3,4,5,6,7). These changes frequently result in decreased vocal clarity and endurance, which may cause older adults to limit social participation or avoid social activities altogether.

Although these severe cases highlight the social consequences of voice disorders, even mild, self-perceived VD without professional diagnosis may substantially limit daily functioning among community dwelling older adults, yet this remains understudied. Severe VD, such as those in patients who underwent laryngectomy, may lead to social withdrawal or reduced participation in group activities. For instance, a survey conducted among patients who underwent laryngectomy revealed that more than 40% of these individuals withdrew from conversation⁸⁾. Similarly, a scoping review identified anxiety regarding communication, social interactions, and being understood, as well as physical challenges requiring vocalization (e.g., speaking loudly or performing certain tasks) following total laryngectomy⁹⁾. These findings demonstrate that vocal limitations can lead to social isolation, increasing the risk of loneliness and depression.

In addition to patients with severe VD, self-perceived VD in older adults represents a major concern. The community-based prevalence of self-perceived VD in older adults ranges from 8.5% to 32.5%^10,11,12,13). However, many individuals attribute hoarseness or vocal weaknesses to “just getting older” and consequently do not seek professional evaluation^{14, 15)}. This reluctance to seek care often results in unrecognized self-perceived VD by healthcare providers, highlighting the need for further study on the impact of self-reported VD on daily functioning. Although a limited number of cross-sectional studies found that participants reporting self-perceived VD engaged in fewer communication-based social activities compared to their counterparts without such complaints¹¹⁾, further longitudinal research is needed to determine whether older adults with self-perceived VD are more likely to avoid social situations over time.

In addition to VD, hearing loss (HL), which is highly prevalent among older populations, affecting approximately two-thirds of individuals over the age of 70¹⁶⁾, negatively impacts communication. Importantly, as the severity of HL increases, older adults have been found to become increasingly socially isolated more, contributing to social and emotional difficulties¹⁷⁾. Older adults with HL are at an increased risk of developing VD, potentially due to altered auditory feedback and vocal strain¹⁸⁾. When VD and HL coexist, individuals experience compounded communication challenges: difficulty projecting a clear, audible voice and limited comprehension of others’ responses. This dual impairment can foster frustration, withdrawal, and a pronounced decline in social participation. Although the association between each perceived condition (VD or HL) and poor psychosocial or functional outcomes in older adults is well established^{15, 19, 20)}, few studies have investigated the interaction of both conditions within the same cohort. One study reported that individuals with both VD and HL are at a greater risk of depression than those with neither condition²¹⁾. Therefore, to guide more effective interventions, it is crucial to understand how self-perceived VD and HL interact and influence social participation.

Given these considerations, there is a clear need to explore the longitudinal impact of self-perceived VD on social participation among older adults. Most previous studies either focused on clinically confirmed dysphonia^{8, 9)} or utilized cross-sectional designs¹¹⁾ and qualitative methods^{19, 22)} of analyses. By analyzing data over time, researchers can more effectively determine the impact of self-perceived VD on social participation. Additionally, investigating how HL exacerbates these patterns will be essential for developing tailored interventions, such as voice therapy, hearing aids, or combined communication strategies, that support social participation in older adults. Therefore, in the current study, we aimed to determine the longitudinal relationship between self-perceived VD and social participation in community-dwelling older Japanese adults, with a particular focus on the potential exacerbating role of HL. We hypothesized that older adults with either self-perceived VD or HL alone will experience poorer social participation than those without such conditions, and that the coexistence of self-perceived VD and HL will pose an even greater risk of declining social participation. Understanding these associations could inform the development of interventions to help maintain social participation among older adults.

PARTICIPANTS AND METHODS

This longitudinal study included community-dwelling older adults who participated in the Kasama Health Checkup for Longevity survey conducted in 2022 and 2023. The survey, initiated in 2009, has been conducted annually in Kasama City²³⁾. The participants were randomly selected from the Basic Resident Registration Network System according to the following inclusion criteria: (i) aged ≥65 years, (ii) residents of Kasama City, and (iii) not receiving long-term care insurance. A flow diagram of the study participants is depicted in Fig. 1. Participants with missing data were excluded from the study, and a total of 134 participants were included in the analysis. This study was approved by the ethics committee of the University of Tsukuba (Tai 30-5) and conducted in accordance with the guidelines of the Declaration of Helsinki. All study participants provided written informed consent to participate in this study.

Fig. 1.

Flow diagram of study participants.

To assess participants’ social activity, the FGO was measured using a question from the Physical Activity Scale for the Elderly (PASE)²⁴⁾, “How much did you walk outside your home in the last 7 days?” Responses were recorded on a 4-point scale: (1) did not walk at all; (2) occasionally walked (1–2 days); (3) occasionally walked (3–4 days); and (4) frequently walked (5–7 days). For analysis, the responses were converted into numerical variables based on the median value of days.

Social interaction was assessed using the Lubben Social Network Scale-6 (LSNS-6)²⁵⁾, which evaluates the size and quality of an individual’s social networks. The LSNS-6 includes three items related to family networks and three items related to non-family (friend) networks. Participants were asked about the number, with each answered based on the number of people they 1) see often, 2) can consult, and 3) can ask for help. The total score ranged from 0 to 30, with higher scores indicating larger social networks. In this study, family and friends scores were used for the analysis.

Self-perceived VD was assessed at baseline using the Voice Handicap Index-10 (VHI-10) self-perceived rating scale, developed in 1997^{26, 27)}. Each of the 10 questions included in VHI-10 queries self-perceived VD in various situations; the frequency of these problems is scored using the 5-point scale: “Not at all, 0”; “A little, 1”; “Sometimes, 2”; “Often, 3”; and “Always, 4”. The total score was calculated, and a score of 5 or higher was defined as “having self-perceived VD”: as described previously²⁸⁾. According to Behlau et al., if the goal of screening is to identify all individuals with VD using the VHI-10, then a more stringent assessment is recommended, with a minimum cut-off value of 5.0 points. This cut-off value was applied in the current study, as it specifically focuses on mild VD.

Self-perceived HL was assessed at baseline using the Hearing Handicap Inventory for the Elderly-Screening Version (HHIE-S), as previously described²⁹⁾. Participants were asked to assess the frequency of hearing-related difficulties. If they used hearing aids, they were instructed to assess their condition while wearing them. The HHIE-S evaluates the frequency of problems caused by HL using a 3-point scale: “no” (0 points), “sometimes” (2 points), or “yes” (4 points). Scores range from 0 (no handicap) to 40 (maximum handicap). A cutoff of 8 or more points has been suggested to identify individuals with HL³⁰⁾.

We collected participants’ data, including sex (male/female), age, socioeconomic status as economic status (considerable hardship, some hardship, normal, somewhat comfortable, very comfortable), years of education (less than 10 years, 10 to 12 years, 13 years or more), subjective health status (very healthy, fairly healthy, not very healthy, not healthy), depressive symptoms, cognitive function, and living alone which were used as covariates in the analysis. Depressive symptoms were assessed using the Geriatric Depression Scale (GDS)³¹⁾, where higher scores indicate worse depressive symptoms. Cognitive function was evaluated using the Five Cog total score. The Five-Cog test was developed by the International Society of Geriatric Psychiatry to test cognitive function and detect age-related cognitive decline. The test comprises five domains: (1) a letter position reference task to evaluate attention; (2) a category cue recall task to evaluate memory; (3) a clock drawing task to evaluate visuospatial function; (4) an animal name listing task to evaluate language; and (5) an analogy task to evaluate reasoning. Higher scores indicated superior performance; the sum of all domains was used as the Five-Cog total score for each participant.

Descriptive data are expressed as mean ± standard deviations (SD) for continuous variables and proportions for categorical variables. To determine the impact of VD and HL on changes in FGO and social interaction after 1 year, a repeated measures two-way ANCOVA was performed. A repeated measures two-way ANCOVA was used to assess changes in FGO and social interactions among participants with VD and HL. To explore the combined association between VD and HL, participants were categorized into three groups: Group 1, no condition (participants with neither VD nor HL); Group 2, single condition (participants with either VD or HL); and Group 3, dual condition (participants with both VD and HL). All analyses were adjusted by sex, age, socioeconomic status, years of education, subjective health status, depressive symptoms, cognitive funtion, and living alone at baseline. Data analyses were performed using IBM SPSS (version 27.0; IBM Corp., Armonk, NY, USA). Statistical significance was set at p<0.05. Results with p-values between 0.05 and 0.10 were considered marginally significant. Simple main effects were examined when interaction effects were significant or marginally significant. At baseline, the groups were comparable in terms of demographic and functional characteristics.

RESULTS

The characteristics of the study participants are presented in Table 1. The average age of participants was 76.8 ± 5.3 years, and 43.8% were male. Among the participants, 25 (18.7%) reported self-perceived VD, whereas 109 (81.3%) did not; 49 (36.6%) reported self-perceived HL, whereas 85 (63.4%) did not.

Table 1. Characteristics of participants in this study.

		Total		Male		Female
		n=134		n=59		n=75
		n	%	n	%	n	%
Mean age, years (Mean ± SD)		76.8 ± 5.3		77.5 ± 4.7		76.2 ± 5.7
Subjective health status
	Very good	22	16.1	13	22.0	9	12.0
	Good	100	74.6	38	64.4	62	82.7
	Not so good	11	8.2	7	11.9	4	5.3
	Not good	1	0.7	1	1.7	0	0.0
Years of education
	Less than high school	11	8.2	5	8.5	6	43.0
	High School graduate	79	59.0	36	61.0	43	57.3
	More than high school	44	32.8	18	30.5	26	34.7
Economic conditions
	Very poor	1	0.7	1	1.7	0	0.0
	Poor	20	14.9	11	18.6	9	12.0
	Normal	82	61.2	32	54.2	50	66.7
	Good	27	20.1	12	20.3	15	20.0
	Very good	4	3.0	3	5.1	1	1.3
Depressive symptoms (GDS), point (Mean ± SD)		2.6 ± 2.6		2.5 ± 2.8		2.7 ± 2.4
5-Cog score, point (Mean ± SD)		85.2 ± 16.7		81.3 ± 16.6		88.3 ± 16.2
Living alone		20	14.9	3	5.1	17
Voice function
	Normal	109	81.3	43	72.9	66	88.0
	Decreased	25	18.7	16	27.1	9	12.0
Hearing
	Normal	85	63.4	34	57.6	51	68.0
	Decreased	49	36.6	25	42.4	24	32.0
Voice × Hearing
	No condition group	78	58.2	29	49.2	49	65.3
	Decreased VD only	7	5.2	5	8.5	2	2.7
	Decreased HL only	31	23.1	14	23.7	17	22.7
	Dual condition group	18	13.4	11	18.6	7	9.3
VHI-10 score, point (Mean ± SD)		2.6 ± 3.9		3.7 ± 4.8		1.7 ± 2.7
HHIE-S score, point (Mean ± SD)		7.4 ± 9.4		8.3 ± 9.8		6.7 ± 9.1

SD: standard deviation; GDS: geriatric depression scale; 5-Cog: five-cognitive test; VD: voice disorders; VHI-10: voice handicap index-10; HL: hearing loss; HHIE-S: hearing handicap inventory for the elderly-screening version

A repeated-measures two-way ANCOVA revealed a significant interaction between time and self-perceived VD for FGO (p=0.047) (Table 2) (Fig. 2). In the self-perceived VD group, FGO significantly decreased from 4.1 to 3.2 days/week (p=0.034), as indicated by the simple main effect analysis. In contrast, no significant interaction effects were observed between time and self-perceived HL for FGO (Table 3). When examining the combined association of self-perceived VD and HL, a two-way repeated-measures ANCOVA demonstrated a marginally significant time-by-group interaction (p=0.091). Simple main effects revealed that only the dual-impairment group showed a significant decrease in FGO over one year (from 4.2 to 3.3 days/week, p=0.036) (Table 4) (Fig. 3). No significant change were observed in the other groups: Group 1: from 3.9 to 4.1 days/week; and Group 2: from 3.6 to 3.2 days/week). No significant interaction effects were observed for the family and friends subscale. In contrast to the FGO results, no significant longitudinal changes were observed in either the family or friend subscales of the LSNS-6 for any exposure group.

Table 2. Relationship between voice disorders and social participation.

		Baseline (day)	After 1 year (day)
Frequency of going outa
	Voice normal, day	3.8 ± 2.2	3.8 ± 2.2
	Voice disorders, day	4.1 ± 2.2	3.2 ± 2.3*
LSNS family subscalea
	Voice normal, point	9.4 ± 2.4	9.1 ± 2.5
	Voice disorders, point	8.8 ± 3.2	8.6 ± 3.2
LSNS friend subscalesa
	Voice normal, point	8.7 ± 3.3	8.4 ± 3.2
	Voice disorders, point	7.5 ± 3.6	7.4 ± 3.5

Each value is presented a mean ± SD. *p<0.05 for the within-group simple main effect of time.

^aAdjusted for sex, age, socioeconomic status, years of education, subjective health status, depressive symptoms, cognitive function, and living alone scores at baseline. VD: voice disorders; LSNS: Lubben Social Network Scale-6.

Fig. 2.

Changes in the frequency of going out according to the presence of self-perceived voice disorder.

Table 3. Relationship between hearing loss and social participation.

		Baseline	After 1 year
Frequency of going outa
	Hearing normal, day	3.9 ± 2.3	4.0 ± 2.2
	Hearing loss, day	3.8 ± 2.0	3.3 ± 2.2
LSNS family subscalea
	Hearing normal, point	9.1 ± 2.6	8.8 ± 2.9
	Hearing loss, point	9.6 ± 2.4	9.4 ± 2.1
LSNS friend subscalesa
	Hearing normal, point	8.7 ± 3.5	8.3 ± 3.5
	Hearing loss, point	8.0 ± 3.1	8.1 ± 2.8

Each value is presented as mean ± SD.

^aAdjusted for sex, age, socioeconomic status, years of education, subjective health status, depressive symptoms, cognitive function, and living alone at baseline. HL: hearing loss; LSNS: Lubben Social Network Scale-6.

Table 4. Combined effect of voice disorders and hearing loss on outings and social participation.

		Baseline	After 1 year
Frequency of going outa
	No condition group, day	3.9 ± 2.2	4.1 ± 2.2
	Single condition group, day	3.6 ± 2.1	3.2 ± 2.2
	Dual condition group, day	4.2 ± 2.0	3.3 ± 2.2*
LSNS family subscalea
	No condition group, point	9.3 ± 2.5	8.9 ± 2.8
	Single condition group, point	9.2 ± 2.5	9.2 ± 2.3
	Dual condition group, point	9.4 ± 2.9	9.0 ± 2.8
LSNS friend subscalesa
	No condition group, point	8.8 ± 3.5	8.3 ± 3.5
	Single condition group, point	8.3 ± 3.0	8.4 ± 2.8
	Dual condition group, point	7.4 ± 3.6	7.3 ± 3.3

Each value is presented a mean ± SD. *p<0.05 for the within-group simple main effect of time.

^aAdjusted for sex, age, socioeconomic status, years of education, subjective health status, depressive symptoms, cognitive function and living alone at baseline. LSNS: Lubben Social Network Scale-6.

Fig. 3.

Changes in the frequency of going out according to combined voice and hearing function status.

DISCUSSION

In this study, we aimed to determine the combined impact of self-perceived VD and HL on FGO and the amount of social interaction among older adults. Analysis of VD and HL in isolation revealed that only VD was associated with a decrease in FGO. Further analysis showed that individuals with both VD and HL experienced a decrease in FGO after 1 year, whereas those with either impairment or no impairment showed no significant change in FGO over the same period.

Consistent with previous studies on self-perceived VD and social life, we found that self-perceived VD led to a decrease in FGO. This has been demonstrated in qualitative studies of older adults^{15, 22)} and cross-sectional studies of both patients⁸⁾ and healthy older adults¹¹⁾. Increased effort and discomfort related to voice have been observed in older adults, resulting in these individuals relying on their spouses for communication and avoiding certain activities, such as social situations and parties^{15, 22)}. In older adults, higher FGO is associated with the maintenance and improvement of activities of daily living and self-efficacy³²⁾. Additionally, lower FGO is associated with an increased risk of mobility decline³³⁾, cognitive deterioration³⁴⁾, and an increased risk of mortality³⁵⁾. Therefore, our results suggest that some older adults in the general population may experience reduced FGO due to mild VD, highlighting the importance of voice care to mitigate these risks.

Interestingly, HL was not significantly associated with FGO, although a marginally significant trend was observed. Age-related HL is typically peripheral in origin, associated with structural and anatomic changes in the cochlea. These HL manifestations affect speech comprehension, especially in noisy environments³⁶⁾. A systematic review conducted in 2020 found that HL is associated with a high risk of loneliness and social isolation^{37, 38)}. This may be partly due to concurrent vestibular dysfunction in the inner ear, alongside HL, which contributes to mobility difficulties and further hinders social interaction³⁹⁾. The present study found a marginally significant association, suggesting that even self-perceived HL may increase risk of future confinement. Although the interaction effects observed for the co-occurring VD and HL were only marginal (p=0.091), the direction of these effects suggests that even mild changes in communication-related functions may relate to decreases of 1 day per week after a year in FGO.

A key finding of this study is that older adults experiencing both VD and HL reduced their FGO by approximately 1 day per week after 1 year. This highlights the potential for these two conditions, when co-occurring, to rapidly intensify social isolation in later life. VD impairs expressive communication, while HL affects receptive abilities, and their combined effect can significantly limit meaningful interaction with others. In addition to restricting social opportunities, this dynamic may also hinder older adults’ access to essential resources and support. Therefore, preventing or addressing at least one of these impairments is critical for mitigating social withdrawal. Although the interaction effect was only marginally significant, our findings may still have clinical relevance. Early detection and targeted interventions could assist older adults in sustaining higher levels of social participation and overall well-being.

Although the amount of social interaction with friends and family was not significantly associated with either HL or VD in the present study, socioemotional selectivity theory suggests that older adults prioritize maintaining the most rewarding social relationships, while gradually letting go of those that are less fulfilling⁴⁰⁾. This pattern suggests that VD and HL may primarily affect broader community engagement rather than intimate social bonds. This may explain why the total number of significant social connections, such as family and friends measured by LSNS-6, was less sensitive to the effects of VD and HL, compared to behavioral measures such as FGO. Consistent with this interpretation, Palmer et al., in their examination of the relationship between HL and social isolation, reported varying effect sizes depending on the types of social indicator used⁴⁰⁾. Future research should aim to identify specific social relationship variables that are likely to be impacted by these disorders, while concurrently investigating multiple indicators related to social relationships such as social network size, the frequency of positive social exchanges, the frequency of negative social exchanges, and the frequency of social participation.

Interventions targeting both VD and HL have been suggested. For example, hearing aids and other aural rehabilitation devices can address HL⁴¹⁾. VD can be managed through posture and neuromuscular adjustments, breathing exercises, and voice therapy. Recent studies have shown that muscle-strengthening training for respiratory muscles, combined with singing exercises, can be effective in improving VD and even enhancing quality of life of patients with VD^{42, 43)}. Our results suggest that these interventions may promote social engagement maintenance among older adults throughout their lives. Furthermore, because VD and HL are interdependent, efforts to prevent VD, along with the implementation of systematic assessment and improvement cycle to prevent the concurrent development of both conditions, may prevent the development of co-occurring disorders.

This study has several strengths, including its longitudinal design and focus on the additive effects of VD and HL on FGO in older adults. However, several limitations should be noted. First, although this was a longitudinal study, the follow-up period of 1 year may have been insufficient to observe significant changes in social network degeneration^{44, 45)}. Future studies should examine network changes over a longer duration. Second, the sample size was relatively small, and participants were recruited from a single city in Japan, limiting the generalizability of the results to other regions or countries. Third, the study relied on limited measures of social life, such as FGO and LSNS-6, which may not fully capture the complexity of social relationships. Incorporating additional measures, such as the quality of social exchanges (e.g., frequency of positive and negative interactions and social participation), would offer a more comprehensive understanding of the impact of VD and HL on social functioning. Fourth, this study was conducted in a single city. Environmental factors (e.g., walkability and proximity to social venues) may influence FGO, and these could not be controlled for in the present analysis. Fifth, the indicators used to assess self-perceived VD and HL were based on subjective reports, which inherently involve bidirectional measurement error. Some participants may perceive impairments that are not clinically significant, whereas others may have clinically meaningful deficits that they do not recognize or are reluctant to acknowledge. Finally, although the indicators used for VD and HL in this study were reasonable, they do not constitute formal diagnoses of voice or hearing impairment.

Funding

This work was supported by the COI STREAM initiative launched in 2013 by MEXT, as well as the COI-NEXT initiative launched in 2020 by MEXT (JPMJPF2017), and in part by Japan Science and Technology Agency SPRING under Grant JPMJSP2124.

Conflict of interest

No conflicts of interest.