Hearing Loss And Functional Decline
Overview
Hearing loss has been recognized as one of the major modifiable risk factors for cognitive decline and dementia. However, studies exploring this connection have yielded mixed results.
This study aimed to assess the long-term relationship between self-reported hearing loss and cognitive/functional abilities in 695 cognitively normal (CN) individuals and 941 participants with mild cognitive impairment (MCI) from the Alzheimer’s Disease Neuroimaging Initiative.
In the cognitively normal group with hearing loss, a notably faster rate of cognitive decline was observed based on the modified preclinical Alzheimer’s cognitive composite. Furthermore, both CN and MCI participants with hearing loss showed significantly faster functional decline according to the functional activities questionnaire (FAQ). However, hearing loss did not significantly increase the likelihood of progression to a more advanced cognitive diagnosis in either group.
These findings align with previous research, showing a strong association between self-reported hearing loss and declines in cognitive and functional abilities over time. However, the study did not find evidence that hearing loss accelerates the transition to more severe cognitive impairment.
Introduction
The global population is aging rapidly, with the percentage of people over 60 expected to increase from 12% to 22% by 2050. Consequently, the prevalence of age-related conditions, such as dementia, is predicted to rise. Alzheimer’s disease (AD) is the most common cause of dementia, with age being the leading risk factor. Currently, about 55 million people worldwide live with dementia, a number projected to reach 152 million by 2050. In the U.S. alone, approximately 6.7 million individuals have AD, and this is expected to grow to 13.8 million by 2060. Identifying modifiable risk factors for AD is therefore essential to developing strategies that could help prevent or delay its onset.
Hearing loss has emerged as a potential risk factor for dementia, affecting approximately one-third of individuals over 65. By 2050, 585 million people over 65 are expected to develop hearing loss, placing them at an increased risk of cognitive decline. Studies have suggested that interventions, such as hearing aids, could mitigate this risk, though much of the evidence relies on self-reported data. These studies are often limited by factors such as demographic differences and inconsistent data on the effectiveness and usage of hearing aids. While a recent randomized controlled trial found no significant effect of hearing interventions on overall cognitive decline, a subgroup analysis indicated that hearing aids might reduce cognitive decline in high-risk older adults over a three-year period.
The relationship between hearing loss and cognitive decline remains complex. Some research links hearing loss with poorer cognitive performance over time and identifies it as a risk factor for accelerated cognitive decline and earlier dementia onset. However, other studies suggest that when adjusted for age and follow-up duration, the association between hearing loss and dementia risk becomes non-significant. Additionally, certain analyses have found no increased dementia risk in individuals with treated hearing loss compared to those without hearing loss. This raises the question of whether hearing loss is an independent risk factor or whether it only heightens dementia risk in conjunction with other factors.
Despite mixed findings, most evidence points to an association between hearing loss and lower scores on cognitive tests, even in areas like visual memory, which are less affected by hearing ability. However, studies on hearing loss as a risk for dementia progression have been inconsistent. Some research supports an increased dementia risk, but only when demographic factors are not controlled for, while others find no such relationship. Understanding whether hearing loss is a modifiable risk factor for cognitive decline is crucial. This study aims to explore the link between self-reported hearing loss and cognitive function using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Specifically, it seeks to evaluate the association between hearing loss and cognitive performance, both at baseline and longitudinally, and to assess whether hearing loss increases the likelihood of cognitive impairment progression.
Method
Data were retrieved from the ADNI database (adni.loni.usc.edu) on September 5, 2022. ADNI, established in 2003 under the leadership of Dr. Michael W. Weiner, aims to determine whether serial MRI, PET scans, biomarkers, and neuropsychological assessments can be used to track the progression of mild cognitive impairment (MCI) and early Alzheimer’s disease (AD).
Participants provided informed consent, and the study protocol received approval from the human research committee at each participating institution. ADNI is a longitudinal study focused on aging, enrolling individuals with a range of cognitive statuses: cognitively normal (CN), subjective memory concerns (SMC), early and late MCI, and AD dementia. Diagnostic criteria for each group are detailed at www.adni-info.org. CN participants showed no cognitive impairment, while SMC participants reported memory issues without measurable cognitive decline. Early MCI (EMCI) participants displayed mild memory impairments, and late MCI (LMCI) participants exhibited more pronounced memory deficits, though both groups maintained functional capabilities. AD dementia participants met clinical criteria for probable AD, as established by NINCDS/ADRDA guidelines.
For the study, EMCI and LMCI participants were combined into a single MCI group, while SMC participants were included in the CN group. Diagnostic status was determined at each visit based on pre-specified criteria and clinical judgment by the site’s primary investigator (PI). Data from ADNI 1, Grand Opportunity (GO), 2, and 3 cohorts were used for analysis.
The study examined the link between hearing loss and cognitive/functional decline. Hearing loss was determined based on self-reports or caregiver accounts provided during initial medical history collection. Participants were classified as having significant hearing loss if they reported using hearing aids or acknowledged hearing difficulties. Objective testing to assess the degree of hearing loss was not performed. Hearing aid use, frequency, and effectiveness were inconsistently reported, so these factors were excluded from the analysis. Participants reporting hearing loss after the screening visit were excluded to minimize reverse causality.
Linear mixed models were employed to assess the relationship between hearing loss and various cognitive outcomes. For cognitively normal participants, the modified Pre-clinical Alzheimer’s Cognitive Composite (mPACC) was used as the primary outcome, while the Alzheimer’s Disease Assessment Scale-cognitive subscale (ADAS-Cog-11) was used for MCI participants. The Functional Activities Questionnaire (FAQ) served as the primary measure of functional performance for both groups. Exploratory outcomes included the Clinical Dementia Rating sum of boxes (CDR-sb), MMSE, Rey Auditory Verbal Learning Test (RAVLT), and other cognitive measures. Covariates in all models included age, sex, education, APOE ɛ4 status, and baseline ADAS-Cog-11 scores.
Further investigation is needed to better understand the impact of hearing loss on the progression of cognitive impairment, including its potential effects across different stages of cognitive decline and its interaction with hearing aid interventions.
Statistical Analysis
Independent analyses were conducted for each baseline cognitive diagnosis category: cognitively normal (CN) and mild cognitive impairment (MCI). Participants were classified into two groups based on their self-reported hearing status: those with hearing loss before screening and those without. Baseline characteristics compared between these groups included age, sex, education level, APOE ɛ4 allele count, and previously documented cognitive outcomes. Continuous variables were analyzed using t-tests, while categorical variables were assessed using chi-square (χ²) tests.
To explore whether hearing loss influenced cognitive decline over time, separate repeated-measures linear mixed models were utilized. The primary explanatory variable was the interaction between hearing loss and time (in months since the baseline visit in the ADNI dataset). Age, sex, education, APOE ɛ4 count, baseline ADAS-Cog-11 scores, and participant ID (as a random intercept) were incorporated as covariates. These models were applied to individuals with baseline CN and MCI diagnoses. The results for the hearing loss and time interaction, including F statistics, parameter estimates, and p-values, were reported. Additional main effects of hearing loss, time, age, sex, education, APOE ɛ4 count, and baseline ADAS-Cog-11 scores are provided in the supplementary materials.
Kaplan-Meier survival curves were used to estimate the progression from CN to MCI or Alzheimer’s dementia, as well as from MCI to Alzheimer’s dementia. Cox proportional hazards models assessed the risk of cognitive conversion, with the same covariates as mentioned earlier. A 95% confidence level (p < 0.05) was used for statistical significance, with no adjustments for multiple comparisons. Statistical analyses were conducted using SPSS Statistics Version 28.0 (IBM Corp.).
Result
The study examined 695 participants who were cognitively normal (CN) at baseline, with 19.3% (134 individuals) reporting hearing loss. Those with hearing loss were older (average age 75.2 vs 72.5 years), more likely to be male (62.7% vs 41.9%), and had more years of education (17.0 vs 16.4 years). Hearing-impaired participants performed worse on cognitive assessments, including ADAS-Cog-11 and FAQ scores, compared to those without hearing loss. Participants without hearing loss were followed for an average of 59 months, while those with hearing loss had an average follow-up of 66 months.
Among 941 participants diagnosed with mild cognitive impairment (MCI) at baseline, 22.4% (211 individuals) reported hearing loss. Similar to the CN group, the majority were white (93.8%) and non-Hispanic (96.6%). MCI participants with hearing loss were older (76.2 vs 72.1 years), predominantly male (76.8% vs 54.4%), and had more education (16.4 vs 15.9 years). They also showed worse baseline cognitive performance on tests such as RAVLT immediate recall and FAQ scores. Cognitive follow-up for MCI participants averaged 49 months for those without hearing loss and 51 months for those with it.
Linear mixed models revealed that CN participants with hearing loss experienced a greater cognitive decline over time, as shown by the mPACC score. While there was no significant interaction between hearing loss and cognitive decline in MCI participants, both CN and MCI groups with hearing loss demonstrated a decline in FAQ scores over time. Exploratory analysis showed that hearing loss was associated with worsening cognitive performance on several measures in both groups, such as the MMSE and RAVLT.
Cox proportional hazard models indicated that hearing loss did not significantly increase the risk of progression to a more impaired diagnostic category. However, age, baseline cognitive performance (ADAS-Cog-11), and the presence of the APOE ɛ4 allele were significant predictors of progression. For CN participants, hearing loss did not significantly increase the risk of progression to MCI, nor did it contribute to the risk of MCI participants progressing to Alzheimer’s disease dementia. Age, baseline cognitive status, and APOE ɛ4 copy number were key risk factors for conversion to more advanced cognitive impairment in both CN and MCI groups.
Conclusion
The study explored how self-reported hearing loss impacts cognitive and functional performance in participants diagnosed with either cognitively normal (CN) or mild cognitive impairment (MCI). It aimed to determine whether hearing loss could predict progression to a more impaired diagnostic category over time. The results indicated that both CN and MCI individuals with reported hearing loss exhibited poorer cognitive and functional outcomes at both baseline and during follow-up assessments compared to those without hearing loss. However, hearing loss was not found to be a predictor of transitioning to a more impaired diagnostic group.
At the study’s outset, participants with self-reported hearing loss tended to be older, predominantly male, and generally more educated. These demographic trends align with existing literature indicating that older adults and males are more prone to hearing loss. Notably, the finding linking hearing loss to higher education levels contradicts some previous studies and may reflect a sampling bias within this cohort, which was primarily composed of non-Hispanic White participants, thereby limiting the broader applicability of the findings.
In terms of cognitive performance, CN individuals with hearing loss showed significant deficits in global cognitive measures (ADAS-Cog-11 and ADNI-Mem), indicating that even cognitively normal individuals with hearing loss are already at a disadvantage. For the MCI group, those with hearing loss performed poorly on the RAVLT immediate recall task, possibly due to challenges in hearing the presented words. Importantly, despite differences in baseline performance, the observed relationship between hearing loss and memory retention (as measured by RAVLT percent forgetting) suggests that hearing loss impacts verbal memory over time, irrespective of initial encoding ability. Additionally, CN participants with hearing loss demonstrated lower functional independence compared to their counterparts without hearing loss, although this trend was not observed in the MCI group.
The study further analyzed longitudinal cognitive and functional data using measures like the mPACC for CN participants and the ADAS-Cog-11 for those with MCI. In the CN cohort, individuals with hearing loss exhibited a decline in global cognitive performance over time, while no significant differences were noted among MCI participants regarding the ADAS-Cog-11 scores. This lack of observable impact in the MCI group may suggest that other factors, such as underlying disease processes, might obscure the cognitive effects of hearing loss once symptoms emerge.
Both CN and MCI participants with self-reported hearing loss consistently scored lower on the FAQ, indicating that hearing impairment adversely affects functional performance over time, independent of cognitive status. Exploratory analyses revealed that memory assessments, particularly the RAVLT percent forgetting and ADNI-Mem scores, were significantly impacted by hearing loss. These findings align with previous research indicating an increased rate of cognitive decline among those with hearing impairment.
Interestingly, the study found no evidence that hearing loss was associated with a higher risk of transitioning to a more impaired diagnostic group, contradicting some earlier studies that identified hearing loss as a possible risk factor for dementia and cognitive decline. This discrepancy suggests that the relationship between hearing loss and cognitive decline may not be straightforward and could be influenced by additional factors.
The study faced several limitations, including a potential lack of power to detect the effects of hearing loss on clinical conversion, particularly given the limited number of participants in the ADNI dataset with extended longitudinal cognitive testing beyond 96 months. Furthermore, the method of assessing hearing loss relied on self-reporting, which is susceptible to bias and may have led to an underestimation of actual hearing loss prevalence. The lack of objective hearing acuity measurements is another notable limitation.
Future research should focus on longer-term studies with larger sample sizes to explore the relationship between hearing loss and cognitive decline. Additionally, collecting more detailed data on hearing loss and hearing aid usage in large-scale studies could provide valuable insights. It is also crucial to examine potential sex-based differences in the association between hearing loss and cognitive outcomes, given the established disparities in hearing loss prevalence.
Given the growing evidence that subjective memory concerns may pose a risk for cognitive decline, the inclusion of participants with such concerns in the CN group limits the ability to assess whether those with both subjective memory issues and hearing loss are at an even greater risk. Future investigations should also consider how hearing loss affects various etiologies and stages of cognitive impairment, including preclinical Alzheimer’s disease.
In summary, the findings from this study enhance the existing literature regarding the negative impact of hearing loss on cognitive and functional performance in both cognitively normal individuals and those with MCI. While the results did not demonstrate an increased risk of transitioning to a more impaired diagnosis associated with hearing loss, they underscore the need for further investigation into the implications of hearing impairment in cognitive health.
