Next Article in Journal
Associations between Neuropsychiatric Symptoms and Alzheimer’s Disease Biomarkers in People with Mild Cognitive Impairment
Previous Article in Journal
The Association between Pesticide Exposure and the Development of Fronto-Temporal Dementia-Cum-Dissociative Disorders: A Review
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Brain Sciences
Volume 13
Issue 8
10.3390/brainsci13081196
brainsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Is Hearing Loss a Risk Factor for Idiopathic Parkinson’s Disease? An English Longitudinal Study of Ageing Analysis

by
Megan Rose Readman
1,2,3,*,†,
Fang Wan
4,†,
Ian Fairman
5,†,
Sally A. Linkenauger
1,†,
Trevor J. Crawford
1,† and
Christopher J. Plack
1,6
1
Department of Psychology, Lancaster University, Lancaster LA1 4YW, UK
2
Department of Primary Care and Mental Health, The University of Liverpool, Liverpool L69 3BX, UK
3
NIHR ARC NWC, Liverpool L7 8XP, UK
4
Department of Mathematics and Statistics, Lancaster University, Lancaster LA1 4YW, UK
5
Public Advisor, Associated with Lancaster University Psychology Department, Lancaster LA1 4YF, UK
6
Manchester Centre for Audiology and Deafness, University of Manchester, Manchester M13 9PL, UK
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Brain Sci. 2023, 13(8), 1196; https://doi.org/10.3390/brainsci13081196
Submission received: 12 July 2023 / Revised: 2 August 2023 / Accepted: 9 August 2023 / Published: 12 August 2023
(This article belongs to the Section Neurodegenerative Diseases)
Review Reports Versions Notes

Abstract

:
Observations that hearing loss is a substantial risk factor for dementia may be accounted for by a common pathology. Mitochondrial oxidative stress and alterations in α-synuclein pathology may be common pathology candidates. Crucially, these candidate pathologies are implicated in Parkinson’s disease (PD). Consequently, hearing loss may be a risk factor for PD. Subsequently, this prospective cohort study of the English Longitudinal Study of Ageing examines whether hearing loss is a risk factor for PD longitudinally. Participants reporting self-reported hearing capabilities and no PD diagnosis prior to entry (n = 14,340) were used. A joint longitudinal and survival model showed that during a median follow up of 10 years (SD = 4.67 years) increased PD risk (p < 0.001), but not self-reported hearing capability (p = 0.402). Additionally, an exploratory binary logistic regression modelling the influence of hearing loss identified using a screening test (n = 4812) on incident PD indicated that neither moderate (p = 0.794), nor moderately severe/severe hearing loss (p = 0.5210), increased PD risk, compared with normal hearing. Whilst discrepancies with prior literature may suggest a neurological link between hearing loss and PD, further large-scale analyses using clinically derived hearing loss are needed.

1. Introduction

Hearing loss is oftentimes considered an inevitable sequela of ageing. However, an accumulation of evidence has revealed that hearing loss is a substantial, potentially modifiable, risk factor for incident all-cause dementia [1]. Specifically, Lin et al. [2] demonstrated that mild hearing loss almost doubles dementia risk, moderate hearing loss triples dementia risk, and severe hearing loss increases dementia risk by almost five times, over ~12 years of follow up.
The underlying mechanism relating hearing loss and dementia remains unclear. The common cause hypothesis suggests that a common pathology affects both the cochlea and ascending auditory pathway (causing hearing loss) and the cortex (causing dementia) [3,4,5]. Mitochondrial oxidative damage and alterations in the production, and aggregation, of α -synuclein are two potential common pathology candidates. Following the rationale of the common cause hypothesis, it may be that hearing loss is a significant risk factor for alternative clinical conditions in which mitochondrial oxidative damage and alterations in the production, and aggregation, of α -synuclein are implicated, such as the progressive neurodegenerative disorder Parkinson’s Disease (PD).
Oxidative stress, a biological state, occurs when the production and accumulation of reactive oxygen species (ROS) within cells outweighs detoxification of these molecules [6]. This state initiates damage to cellular macromolecules, including mitochondrial membranes, respiratory chain proteins, and nuclear and mitochondrial DNA [7], which can eventually trigger upstream cell apoptosis [8]. Evidence indicates that mitochondrial oxidative stress is implicated in dementia pathogenesis [9]. For example, increased levels of DNA strand breaks [10,11], and brain lesions [12], characteristic of oxidative stress, have been observed in Alzheimer’s disease. Similarly, mitochondrial oxidative damage in the cochlea is also implicated in the pathogenesis of acquired hearing loss [13,14]. Specifically, mitochondrial DNA alterations, indicative of oxidative stress, in the cochlear tissue are more frequent in patients with age-related hearing loss compared to those with normal hearing [15]. Furthermore, genetic variations in certain antioxidant defence genes have been associated with increased risk of age-related hearing loss [16,17,18].
PD is pathologically hallmarked by accelerated dopaminergic neuronal cell death, particularly within the substantia nigra of the basal ganglia [19]. Whilst the mechanism underlying this is not yet elucidated, several patomechanisms have been hypothesised. One of these hypotheses focuses on the role of chronic neuroinflammation, microglial activation and the subsequent oxidative stress [20]. Indeed, evidence suggests that regular non-steroidal anti-inflammatory drug (NSAID) users are at reduced risk (~36–55 percent reduced) of PD [20]. Moreover, significantly elevated inflammation levels are observed in the blood, identified through neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratios, and the cerebrospinal fluid of PD patients relative to controls [21]. Importantly, it is postulated that activated microglia produce large amounts of reactive oxygen species (superoxide radicals) thus implicating oxidative damage in PD [22]. Supporting this assertion, consumption of MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine), which increases ROS production and leads to a sustained inflammatory reaction, can induce Parkinsonism [23]. Comparably, elevated circulating oxidative stress markers (e.g., ferritin, 8-OhdG and nitrite) and decreased antioxidant levels (e.g., catalase, uric acid, glutathione) have been observed in PD [24,25].
Another potential common pathological candidate, for the relation between hearing loss and dementia, is alterations in the production, and aggregation, of α -synuclein. Natively, α -synuclein is unfolded. However, in response to specific environmental factors, such as toxins and inflammation, α -synuclein folds into monomers, polymers, and oligomers [26]. Notably, folded α -synuclein are major components of Lewy bodies [27].
Importantly, α -synuclein Lewy body pathology has been detected in over 50 percent of sporadic [28,29] and ~60 percent of familial [30], Alzheimer’s disease cases at autopsy. Similarly, elevated levels of soluble α -synuclein are observed in the cerebral spinal fluid of patients with Alzheimer’s disease [31] and mild cognitive impairment [32,33] relative to controls. Regarding hearing loss, Park et al. [34] observed weaker efferent nerve and cochlear homogenate α -synuclein expression in early-onset hearing loss, compared to late-onset hearing loss mouse models. Furthermore, Akil et al. [35] found significantly elevated auditory brain stem thresholds in α -synuclein overexpression transgenic mice compared to wild-type mice, implicating α -synuclein overexpression in hearing loss.
Critically, in addition to accelerated dopaminergic neuronal degeneration, PD is also neuropathologically hallmarked by intracellular Lewy body inclusions and Lewy neurites [19]. Moreover, mutations (e.g., PARK1 and PARK4) in the α -synuclein gene (SNCA) are accountable for autosomal dominant PD [36], and variations in the SNCA gene appear to be one of the strongest risk factors in the development of sporadic PD [37]. Subsequently, α -synuclein may contribute to the neuropathology of PD [38].
Thus far, mitochondrial oxidative damage and alterations in α -synuclein aggregation have been presented as dissociable pathologies of hearing loss and dementia. However, these pathologies may form a positive feedback loop in which either event may trigger a self-perpetuating process [39,40,41]. Given that mitochondrial oxidative stress and alterations in α -synuclein production and aggregation are central to PD pathogenesis [42], following the rationale of the common cause hypothesis [3], we may hypothesise that hearing loss will be intricately related to PD.
Consideration of the relation between essential tremor and hearing loss may further support the assumption that hearing loss will be intricately related to PD. Evidence suggests that patients with essential tremor, which can be defined as “bilateral upper extremity action tremor” [43], are at an elevated risk of PD [44]. Specifically, some analyses report that patients with essential tremor are up to four times more likely to develop PD than people without essential tremor [45]. Importantly, hearing loss appears to also be present in essential tremor patients [46]. Some studies have shown that essential tremor patients have significantly elevated pure tone audiometry thresholds at high but not low frequencies [47], and others have observed significantly elevated pure tone audiometry thresholds at low but not high frequencies [48]. Given that essential tremor is a risk factor for PD, and patients with essential tremor are likely to experience hearing loss, we may predict that hearing loss is likely to proceed PD.
Indeed, a meta-analytical review concluded that sensorineural hearing loss and cochlear impairment are more severe in PD patients compared to age-matched controls [49]. Moreover, asymptomatic hearing impairment (when indicators of hearing loss are observed without self-reported hearing difficulties) is significantly higher in PD patients than age-matched controls [50,51]. Consequently, altered audiological function may be related to PD [52]. However, whether hearing loss results from complex sensory processing impairments during PD, or whether hearing loss antedates PD diagnosis, and is therefore a risk factor for PD, remains unclear.
Recent prospective studies indicate that hearing loss may antedate, and be a risk factor for PD. For example, Schrag et al. [53] found that hearing loss, up to 5 years pre-diagnosis, was more prevalent amongst those who later developed PD than controls. Moreover, Lai et al. [54] found that, over a five-year follow-up, hearing loss increases the risk of incident PD by 1.5-fold. Similarly, Simonet et al. [55] observed that hearing loss increases PD risk by 1.6-fold over <2 years follow-up. These studies implicate hearing loss as a risk factor for subsequent PD diagnosis; however, further research is required to substantiate these claims within alternative populations (i.e., using different databases/cohorts). Moreover, these studies use clinical hearing loss diagnosis (i.e., a documented medical diagnosis) as a dependent measure. As such, whether hearing loss at different neurophysiological levels, using different measures of hearing loss, antedates PD diagnosis, and so may be a risk factor for PD, remains unclear. This study aims, therefore, to determine whether hearing loss is a risk factor for later PD diagnosis using data from the English longitudinal study of ageing (ELSA) [56].

2. Materials and Methods

This study was pre-registered on the Open Science Framework (OSF; https://doi.org/10.17605/OSF.IO/MJGQ6, accessed on 1 February 2023), and the data analysed in this study are openly available in the UK Data Service at https://doi.org/10.5255/UKDA-SN-5050-25, accessed on 18 November 2022. The OSF page for this project includes details regarding the variables analysed, planned statistical analyses, and the data analysis code book. We summarise the implementation below. This study deviated from the pre-registration only in the number of covariates embedded within both the primary and exploratory analyses and the exclusion criteria for missing data. Please refer to Appendix A for full justification for these deviations.

2.1. Study Population

2.1.1. Primary Analysis

The ELSA is an ongoing longitudinal prospective cohort study of adults living in private households in England [56]. Data collection follows a longitudinal design with repeated measures of core variables occurring biannually over numerous ‘waves’. The present study utilised the full dataset, including core participants and younger and older responding relatives, from waves 1–9 (see Appendix B for further sampling details). Participants were excluded if they:
1.
Reported a diagnosis of PD at wave 1;
2.
Appeared only in one wave;
3.
Were missing data for self-reported hearing capabilities, age, or PD diagnosis variables.
The final analytical sample was n =14,340 (6466 males; 7874 females).

2.1.2. Post-Hoc Exploratory Analysis

We used HearCheckTM Screener data from wave 7 and incident PD data from wave 9 of ELSA. Participants reporting a PD diagnosis at wave 7 who did not complete the HearCheckTM Screener test, or who did not have age data available, were excluded. The final analytical sample was n = 4812 (2065 males, 2747 females).

2.2. Outcome Measure

2.2.1. Primary Analysis

Participants disclosed whether they had a diagnosis of PD at all waves. PD diagnosis was confirmed at subsequent waves.

2.2.2. Post-Hoc Exploratory Analysis

Participant reported PD in wave 9.

2.3. Exposure Measure

2.3.1. Primary Analysis

Participants self-reported hearing capabilities at each wave by indicating on a five-point Likert scale “Is your hearing [using a hearing aid as usual] (1) excellent, (2) very good, (3) good, (4) fair, or (5) poor”.

2.3.2. Post-Hoc Exploratory Analysis

The HearCheckTM screener generates a series of six tones: three mid-frequency tones of decreasing volume at 1 kHz (55, 34, and 20 dB) and three high-frequency tones of decreasing volume at 3 kHz (at 75, 55, and 35 dB). Participants indicate when they hear the tone by raising their finger. Audibility testing occurs separately for each ear. Participants are required to remove any hearing aid(s) prior to completion of the screening test, thus the obtained results reflect hearing capabilities in the absence of corrective hearing devices. Evidence indicates that hearing loss, defined as >35 dB at 3 kHz in the better-hearing ear, is a robust and justifiable cut off criterion at which intervention is beneficial [57,58]. Subsequently, hearing loss was defined according to this criterion. The resulting hearing loss variable was further subdivided into two mutually exclusive categories: (1) ‘moderate hearing loss’: >35 dB to 54 dB (tone not heard at 34 dB HL but heard at 55 dB and at 75 dB) and (2) ‘moderately severe or severe loss’: >55 dB HL (tone not heard at 35 dB HL and at 55 dB HL, but the tone may, or may not, have been heard at 75 dB HL).

2.4. Covariates

Age is the biggest risk factor for developing PD [59] and hearing loss [60]. Hence, we controlled for age, as a continuous variable, in all analyses. Age at entry into ELSA is used as a baseline covariate, while time spent in ELSA (derived from wave number) is used as the timescale for time to diagnosis of PD and the longitudinal HL outcomes.

2.5. Data Analysis

2.5.1. Primary Analysis

Joint longitudinal and survival models consider the potential dependency between the longitudinal and survival data [61], assessing the impact a longitudinal covariate, measured with error, has on the time to an event of interest [62]. Hence, a joint longitudinal and survival model was fitted to assess the relation between self-reported hearing loss and subsequent PD, with age as a covariate.
Model structure. There are two components to a joint model: a longitudinal component and a time-to-event survival component. The longitudinal and survival components of the model were joined through a trajectory function. (see Appendix C for full joint model mathematical formulae).
Longitudinal component. The longitudinal component consisted of a continuation ratio model for the ordinal hearing loss variable, which assumes a linear model with random effects for the log odds of being in the kth category, conditional on being in the kth category or higher.
Survival component. The survival component, and time to PD diagnosis, of the joint model consisted of a Weibull model.

2.5.2. Post-Hoc Exploratory Analysis

Self-reported hearing capabilities are oftentimes influenced by non-auditory factors, including demographic and socioeconomic characteristics (e.g., biological sex and educational attainment [63] and medical history [64]). Indeed, almost one-third of participants with screening-test-identified hearing loss in ELSA wave 7 went undetected by self-report measures [65], indicating a low concordance between self-reported and screening test hearing capabilities. The wave 7 ELSA screening test hearing loss measure is obtained from the Siemens HearCheckTM Screener. The HearCheckTM Screener yields a sensitivity of 0.89 and specificity of 0.86 [66], thus may constitute a more reliable measure of hearing loss. Hence, we conducted a further exploratory analysis to determine whether screening-test-derived hearing loss is a risk factor for subsequent PD diagnosis. The HearCheckTM screening test is not conducted at every wave, rather it was only conducted at wave 7. As such, only one HearCheckTM Screener data point is currently available per participant, thus not permitting a longitudinal analysis. Therefore, to determine whether screening-test-derived hearing loss is a risk factor for subsequent PD diagnosis, a case-control binary logistic regression was conducted.

3. Results

3.1. Primary Analysis

3.1.1. Sample Demographics

During a median follow up of 10 years (SD = 4.67 years), 151 cases of incident PD were reported in the analysed sample (n = 14,340), with 89 of these being male.

3.1.2. Joint Longitudinal and Survival Model

A notable proportion of participants’ self-reported hearing capabilities responses remained stable between two wave points (e.g., wave 7 to wave 8), with the average proportion (across all nine waves) of participants reporting the same hearing capabilities at two consecutive waves being 0.46 for excellent, 0.43 for very good, 0.49 for good, 0.46 for fair and 0.42 for poor. A small number of participants reported their hearing capabilities at a given wave to be better than the prior wave. This proportion marginally increased as self-reported hearing capabilities decreased. That is, a marginally larger proportion of participants reported their hearing capabilities as being better at a subsequent wave if they reported their hearing capabilities to be fair or poor at the prior wave, than participants who reported their hearing to be good, very good, or excellent at the prior wave. Across all waves, a substantial proportion of participants reported their hearing capabilities at a given wave to be worse than the prior wave. Of these participants, most reported their hearing capabilities to be one rating scale point lower than the prior wave (i.e., from excellent at wave 1 to good at wave 2). However, some participants did report a reduction of two or three rating scale points. A full breakdown of the stability of self-reported hearing capabilities can be found in the Supplementary Materials.
Longitudinal outcome. Both age at entry into ELSA (p < 0.001) and length of time since entering the study (time in study, p < 0.001) were significantly associated with self-reported hearing loss. Specifically, for each unit increase in either age or time since entering the study, the odds of being in a better self-reported hearing category decreases by 5.7 percent (95 CI: 5.4–6.0).
Survival outcome. Concerning incident PD diagnosis, as age increases, the likelihood of PD diagnosis increases (p < 0.001). However, self-reported hearing capability was not related significantly to subsequent PD diagnosis (p = 0.402) (see Table 1).

3.2. Post-Hoc Exploratory Analysis

3.2.1. Sample Demographics

Over the ~4 year follow up (the duration between completion of wave 7 and 9), 24 (13 male, 11 female) cases of incident PD were reported in the analysed sample (n = 4812), with the mean age of people with PD being 74.67 (6.81).

3.2.2. Binary Regression

In accordance with the HearCheckTM screener, 3575 participants were categorised as having ‘normal hearing’, 850 participants were categorised as having ‘moderate hearing loss’, and 387 participants were categorised as having ‘moderately severe/severe hearing loss’. Generally, participants with greater hearing loss were more likely to be older and male. See Table 2 for the distribution of demographic characteristics of participants categorised by HearCheckTM screener derived HL.
The assumption of linearity for the age variable was assumed by observing conditional density plots and using the Box Tidwell test (z = 0.316, p = 0.752). After controlling for age, compared with normal hearing, neither moderate hearing loss [OR = 0.87 (95 CI: 0.27–2.41), p = 0.794], nor moderately severe/severe hearing loss [OR = 1.45 (95 CI: 0.43–4.31), p = 0.521], increased risk of incident PD over a four-year follow-up.

4. Discussion

The present study examined whether hearing loss is a risk factor for incident PD. Regarding the incidence of PD, we observed that self-reported hearing capability was not a significant risk factor for incident PD over ~10 years of follow up. Similarly, HearCheckTM screener-derived hearing loss was not a significant risk factor for incident PD over ~4 years. These observations may indicate that interventions for hearing loss (i.e., hearing aids or cochlear implants) may not affect PD progression. However, due to fundamental limitations in the dataset analyses (e.g., lack of data regarding hearing aid use and PD subcategories further research is needed to confirm this hypothesis. In terms of the longitudinal outcome (self-reported hearing capabilities) we observed that as age and length of time in the study increased, self-reported hearing capabilities decreased. Age-related hearing loss is characterised by degenerative pathology in the cochlear hair cells, supporting cells, and auditory nerve endings [67], and age is the largest risk factor for hearing loss [60]. Thus, this observation is to be expected. However, as it is not possible to manipulate increases in chronological age, it is implausible for this factor to be a clinical intervention target.
Although the present study did not observe a relation between hearing loss and incident PD, some previous literature suggests that hearing loss increases the future risk of incident PD by 1.14–1.6 fold [53,54,55]. Critically, this prior literature has relied upon clinically diagnosed hearing loss as the exposure measure. Within the localities in which these studies recruited participants (Taiwanese, English, and German populations), the clinical diagnosis of hearing loss relies upon pure tone audiometry assessments [68,69]. Previous literature has shown that the concordance between pure tone audiometry and self-reported hearing capabilities is low [65]. Hence, this discrepancy in hearing loss measures may explain the lack of concordance in the results.
Exploratory analyses also found that screening-test-derived hearing loss was not a significant risk factor for incident PD. Yet, whilst the HearCheckTM screener constitutes a substantially more objective hearing loss measure than self-report questionnaires, the accuracy and sensitivity of the HearCheckTM are, at the lowest, 83 percent [66]. Consequently, discrepancies with prior research could result from use of a screening test measure with lower accuracy and sensitivity. Crucially, these exploratory findings should be treated with caution due to the small number of incident PD cases reported (n = 24). Additional larger-scale cohort analyses that draw on a more reliable diagnosis of hearing loss (e.g., pure tone audiometry) are required to further elucidate whether hearing loss is a risk factor for incident PD.
The discrepancies between the current findings and those of previous research could speak to the mechanism underlying hearing loss and antedating PD motor manifestations. Pure tone audiometry assessment is thought to depend primarily upon the health of the outer hair cells, and additionally cochlear transduction by the inner hair cells [70]. Although pure tone audiometry can be influenced by psychophysiological factors (e.g., patient attention) and demographic factors (e.g., biological sex) [71], it is largely assumed that pure tone audiometry outcomes primarily derive from peripheral sensory processes. In contrast, self-reported hearing capabilities are heavily influenced by non-auditory factors including demographic (e.g., biological sex, age), socioeconomic (e.g., educational level, occupation), and lifestyle factors (e.g., tobacco and alcohol consumption and physical inactivity [65]), as well as social stigma [72]. Hence, self-reported hearing loss may reflect higher-order subjective processes. Given that audiometrically derived [53,54,55], but not self-reported, hearing capabilities appear to be a risk factor for incident PD, this may suggest that hearing loss and PD are potentially related at a neurological level. This is consistent, potentially, with the speculations regarding the molecular basis of the common cause hypothesis.
Importantly, both the present study and previous research [53,54,55] have exclusively focused on whether hearing loss is a substantial risk factor for incident PD. Parkinsonism is an umbrella term used to describe a group of conditions clinically characterised by motor impairments including tremors, slowed movement, muscle rigidity, gait alterations, and postural instability [73]. Idiopathic PD is the most common Parkinsonism disorder, accounting for 85 percent of neurodegenerative cases. The remaining 15 percent of cases are attributed to atypical Parkinsonian disorders including Progressive Supranuclear Palsy (PSP), Multiple Systems Atrophy (MSA), Dementia with Lewy Bodies (DLB), and Corticobasal Degeneration (CBD). On a neuropathological level, atypical Parkinsonism disorders share several similarities with PD. For example, MSA and DLB, akin to PD, are α -synucleinopathies [74]. Moreover, neuroinflammation and oxidative stress have been implicated as patomechanisms in MSA [75,76], PSP [77,78] and CBD [79,80] (note that the evidence implicating oxidative stress in CBD is less extensive). However, whilst some studies have found DLB to be characterised by a lack of neuroinflammation [81], alternative studies have observed oxidative damage in DLB patients [82].
The present study observed that self-reported and HearCheckTM screener-derived hearing loss are not risk factors for incident PD. However, if clinically diagnosed hearing loss is a risk factor for incident PD [53,54,55], it may be that, due to the neuropathological similarities between PD, PSP, MSA, DLB, and CBD, clinically diagnosed hearing loss will also be a risk factor for other Parkinsonism disorders. Alternatively, with respect to the common cause hypothesis, it may perhaps be that hearing loss is a risk factor for some, but not all, Parkinsonism disorders. Specifically, if hearing loss is related to neurodegenerative disorders through a neuroinflammation and oxidative stress common cause, we may anticipate that hearing loss will also be a risk factor for PSP, MSA, DLB, and CBD. However, if the link is driven by alterations in the production, and aggregation, of α -synuclein, then we would perhaps expect hearing loss to be a risk factor for MSA and DLB, but not PSP or CBD (as both are tauopathies [74]).
Although hearing loss may play a small role in the clinical manifestation of Parkinsonism disorders, observations of a differential influence of hearing loss on symptom onset across Parkinsonism disorders may be useful for clinicians to consider when differentiating such symptomatically similar conditions. Furthermore, ascertaining the relation between hearing loss and multiple Parkinsonism disorders may provide invaluable insights for the differential clinical management of Parkinsonism disorders. Therefore, the predictions made here should be investigated formally.
Within this study, PD was indicated by self-disclosure of a clinical diagnosis. In the UK, in accordance with UK Parkinson’s Disease Society Brain Bank guidelines, a probable diagnosis of PD is provided if any given individual presents with bradykinesia (slowed movement) and at least one of the following: muscular rigidity, rest tremor, or postural instability [83]. Whilst diagnostic criteria emphasise motor manifestations, cognitive disturbances are increasingly recognised as commonplace in PD. Specifically, 30 percent of people with PD will experience dementia [84]. At a pathological level, Parkinson’s dementia is oftentimes coupled with Alzheimer’s disease-related pathologies [85]. Given the evidence that hearing loss is a risk factor for incident all-cause dementia [1], it may be that hearing loss forms a risk factor for the occurrence and severity of cognitive impairment in PD, rather than overall diagnosis.
Clinical heterogeneity is well recognised in PD. Such heterogeneity presents a major barrier in terms of understanding disease mechanisms, developing treatments, and clinical management. Consequently, a set of clinical criteria for PD subtype categorization has been developed. The most frequently referred to subtyping system categorises patients into three subtypes: (1) mild-motor predominant, (2) diffuse malignant, and (3) intermediate, with categorization being based on motor impairment, cognitive function, rapid eye movement behaviour disorder, and autonomic symptoms [86]. Given that specific motor and non-motor symptoms are more strongly associated with some subcategories than others, it may be that hearing loss is a stronger risk factor for PD onset for certain subtypes over others. As previously described, PD diagnosis is a binary self-reported variable within the ELSA dataset. Furthermore, participants are not required to disclose further clinical information, nor are they required to complete PD symptomology assessments (e.g., the Unified Parkinson’s Disease Rating Scale [87]). Consequently, it is not possible to subtype the PD patients within the ELSA dataset, hence rendering it impossible to determine the influence of PD subtype on the relation between hearing loss and incident PD. Further studies that collect sufficient clinical data to allow for subtyping should be conducted.
It is important to also consider the potential role of addressed versus unaddressed hearing loss (i.e., the use of hearing aids) on symptom onset and the results obtained here. When accounting for the relation between hearing loss and dementia, it is increasingly acknowledged that prolonged auditory deprivation, due to hearing loss, may give rise to chronic cortical reorganisation, which hinders cognitive processes in favour of auditory perception [3,88]. Although widely debated, due to the potential influence of sociodemographic factors such as educational level and wealth [89], some evidence also suggests that correction of hearing loss, through hearing aids, may be associated with better cognition and a reduction in cognitive change [90,91]. With respect to the relation between hearing loss and PD, it may be that prolonged auditory deprivation, due to hearing loss, gives rise to chronic cortical reorganisation that favours auditory perception to the detriment of motor control, and such an effect may be in some way mitigated by correction of hearing loss (e.g., the use of hearing aids or cochlear implants). Hence, it may be important for the use of hearing aids to be controlled within analyses of the relation between hearing loss and PD. However, the nature of the dataset analysed here renders it impossible for the potential influence of addressed versus unaddressed hearing loss on symptom onset to be accounted for within this analysis. Specifically, participants are only asked to disclose whether they use a hearing aid in wave 7 (whilst completing the HearCheck Screening test). Consequently, hearing aid use cannot be added as a longitudinal covariate to the joint longitudinal and survival analysis. Therefore, these findings should be treated with a degree of caution, and further studies that control for the use of hearing aids should be conducted.
This study is not without limitations. First, incident PD cases within the final analytical sample were low (n = 151). However, it is unlikely that this substantially biased our findings. Specifically, in 2018, the incidence rate of PD in the English population was 0.22 percent [92]. Within this study, the incidence rate of PD was 1.05 percent. Thus, although the raw number of incident PD cases was low, this is reflective of the overall incidence rate of PD in the general population. Second, caution must be applied when generalising from the results of this study. The ELSA is a volunteer cohort [56]. Although the sample aims to be representative, there is a greater proportion of females, people from managerial/technical and skilled manual social classes, and ethnic majority groups [56,93]. Therefore, further studies using alternative, perhaps more representative, samples are required. Finally, due to the lack of repeat data regarding the use of hearing aids, it is impossible for the potential influence of hearing loss treatment (i.e., hearing aids or cochlear implants) on symptom onset to be accounted for. Therefore, further studies that collect additional data regarding hearing loss treatment are required.

5. Conclusions

To conclude, the present study observed that self-reported hearing loss is not a substantial risk factor for incident PD. However, sample limitations necessitate further analyses in alternative populations to substantiate this finding. Should findings be replicated, treatment of hearing loss may not substantially reduce the risk of developing PD. Additionally, exploratory analyses indicated that more objectively derived hearing loss is also not a risk factor for incident PD. However, given the small number of incident PD cases in this sample, further large-scale, planned, analyses are needed to corroborate this conclusion.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/brainsci13081196/s1, Table S1: Change in self-reported hearing capabilities across waves.

Author Contributions

M.R.R.: Funding acquisition, Conceptualization, Methodology, Formal Analysis, Data Curation, Writing—Original Draft; F.W.: Methodology, Formal Analysis, Data Curation, Writing—Review & Editing; I.F.: Conceptualization, Writing—Review & Editing; S.A.L.: Conceptualization, Writing—Review & Editing; T.J.C.: Conceptualization, Writing—Review & Editing; C.J.P.: Funding acquisition, Conceptualization, Methodology, Writing—Review & Editing, Supervision. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Economic Social Research Council [grant number ES/X004082/1]. M.R.R receives support from the NIHR Applied Research Collaboration ARC North West Coast and Alzheimer’s Society and are funded through a Post-Doctoral Fellowship. The views expressed are those of the authors and not necessarily those of the funders, NHS or Department of Health and Social Care.

Institutional Review Board Statement

This secondary data analysis received favourable ethical opinion from the Faculty of Science and Technology at Lancaster University. Moreover, the English Longitudinal Study of Ageing received favourable ethical approval from the National Research Ethics Service (London Multicentre Research Ethics Committee (MREC/01/2/91)).

Informed Consent Statement

As this study incorporated secondary data analysis, no new data were collected. However, informed consent was obtained from all participants involved in the English Longitudinal Study of Ageing at the time of participation. Further details regarding the informed consent procedure can be found elsewhere, e.g., [74].

Data Availability Statement

The data analysed in this study are openly available in the UK Data Service at https://doi.org/10.5255/UKDA-SN-5050-25, accessed on 11 July 2023.

Acknowledgments

The authors would like to thank all participants who have contributed to the English Longitudinal Study of Ageing.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A. Deviations from Pre-Registration

Below, we provide point-by-point justification for the deviation from the pre-registration analysis plan that occurred within the final analysis.
1.
The planned analyses detailed that “Age, sex, smoking, and alcohol will be controlled for as covariates” in all analyses. However, in the final reported analyses, only age was fitted as a covariate.
Within the current body of literature, there is debate regarding the advantages and disadvantages of embedded covariates to any given analysis. Indeed, adjusting for baseline, highly prognostic, covariates can increase the power to detect a true effect without inflation of a type 1 error [94]. However, any given dataset can tolerate only a limited number of parameters; when too many covariates are embedded within an analysis, the issue of overfitting occurs. Overfitting is a phenomenon in which a given model fits too well to the given data, and hence the modelled effect does not truly exist in the population [95]. That is, the model is fitted not only to the systematic variability but also the random variability in the given dataset, and so will not replicate within an alternative dataset [96]. As such, within the statistical literature, there is a suggestion that fitting variance components to mixed-effects models should occur in step-by-step fashion driven by theoretical considerations, and the resulting model should be the simplest model that effectively explains the given data [97], i.e., the strongest prognostic covariate should be fitted first and then, should the statistical analyses deem it appropriate, the second strongest prognostic covariate should be added, etc. Age is the largest prognostic factor for developing both PD [59] and hearing loss [60]. Hence, age was added as a covariate first. The subsequent age controlled joint longitudinal survival model and binary case-control regression indicated that the exposure variables of interest, self-reported hearing capabilities and objective hearing loss, respectively, were not related to subsequent PD diagnosis (p = 0.402, p = 0.794 (moderate hearing loss), p = 0.521 (moderately severe/severe hearing loss)). Given that the exposure variables of interest was not significant in the given model, we deemed it inappropriate to embed further covariates and rather maintain the principles of parsimony and retain the simplest appropriate statistical model;
2.
Regarding missing data, the planned analysis detailed “as we are interested in whether hearing loss is a risk factor for the onset of Parkinson’s disease, we will exclude all participants who have a known diagnosis of Parkinson’s at Wave 1. Participants whose responses are coded as 0 = Refusal, −8.0 = Do not Know, −1.0 = Not applicable, on the hearing (hehear) measure will also be excluded from analysis.”. Within both the primary and exploratory analyses, age was fitted as a covariate. Thus, we also excluded any participants who had missing data for the age variable.

Appendix B. English Longitudinal Study of Ageing (ELSA) Sampling Information

Table A1. Number of Interviews by Waves in the ELSA.
Table A1. Number of Interviews by Waves in the ELSA.
Fieldwork PeriodTotal Interviews
Wave 1March 2002–March 200312,099
Wave 2June 2004–July 20059432
Wave 3May 2006–August 20079771
Wave 4May 2008–July 200911,050
Wave 5June 2010–July 201110,274
Wave 6May 2012–June 201310,601
Wave 7June 2014–May 20159666 (Nurse Interviews: Nurse Visit—8054)
Wave 8May 2016–June 20178445
Wave 9May 2018–June 20198736

Appendix C. Joint Longitudinal and Survival Model Analysis Mathematical Formulae

Appendix C.1. Survival Component

h i ( + 1 m i ( t ) ) = H o ( t ) exp ( γ T W 2 − t α m i ( t ) )
where α mi ( t ) is the direct effect of hearing loss on the time to PD diagnosis, and γ T W2 measures the association between hearing loss and time to PD diagnosis.

Appendix C.2. Longitudinal Component

The longitudinal component consisted of a continuous ratio mixed effect model.
Y i ( t ) = m i ( t ) + E i ( t )
where mi ( t ) is the true hearing capabilities, but we only observe this with error Ei ( t ) .
Joint Longitudinal and Survival Model: log { Pr ( y ij = K | y ij ≥ K ) 1 − P ( y ij = K | y ij ≥ K ) } = αk + γijt β + Zjtbi.
For K = 1, 2, 3, 4, i = participant, ij = jth observation from participant i.
P(HLij = 1|bi) = logit−1 [ β 01 + bij + β ageAgei + ( β t + Bij)TimeJ] = logit−1 (1.3569 + bij− 0.0584 × Age + (bij − 0.0582) × TimeJ).
Hearing Capabilities ≥ 2: P(HLij = 2|bi) = P(HLij = 2|HLij > 1, bi) = logit−1 [ β 02 + B1i + β AgeAgei + ( β t + β 2i)TimeJ] × [1 − P(HLij = 1|bi)].
β 02 = Intercept + Cohort (≥2) = 1.3569 + 1.8124.
Hearing Capabilities ≥ 3: P(HLij = 3|bi) = P(HLij = 3|HLij > 2, bi) = P(HLij = 3|HLJ > 2, bi) × [1 − P(HLJ = 2| HLJ > 1, Bi)] = logit−1 [ β 03 + B1i + β AgeAgei + ( β t + β 2i)TimeJ] × [1 − P(HLij = 2|HLJ > bi)] × [1 − P (HLJ = 1, Bi)].
β 03 = Intercept + Cohort (≥3) = 1.3569 + 4.1794.
Hearing Capabilities ≥ 4: P(HLij = 4|bi) = P(HLij = 4|HLij > 3, bi) = P(HLij = 4|HLJ > 3, bi) × [1 − P(HLJ = 3| HLJ > 2, Bi)] = logit−1 [ β 03 + B1i + β AgeAgei + ( β t + β 2i)TimeJ] × [1 − P(HLij = 3|HLJ > bi)] × [1 − P(HLJ = 2, Bi)].
β 03 = Intercept + Cohort (≥4) = 1.3569 + 6.3381.

References

  1. Loughrey, D.G.; Kelly, M.E.; Kelley, G.A.; Brennan, S.; Lawlor, B.A. Association of age-related hearing loss with cognitive function, cognitive impairment, and dementia: A systematic review and meta-analysis. JAMA Otolaryngol. Head Neck Surg. 2018, 144, 115–126. [Google Scholar] [CrossRef] [PubMed]
  2. Lin, F.R.; Metter, E.J.; O’Brien, R.J.; Resnick, S.M.; Zonderman, A.B.; Ferrucci, L. Hearing loss and incident dementia. Arch. Neurol. 2011, 68, 214–220. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. Griffiths, T.D.; Lad, M.; Kumar, S.; Holmes, E.; McMurray, B.; Maguire, E.A.; Billig, A.J.; Sedley, W. How can hearing loss cause dementia? Neuron 2020, 108, 401–412. [Google Scholar] [CrossRef] [PubMed]
  4. Nadhimi, Y.; Llano, D.A. Does hearing loss lead to dementia? A review of the literature. Hear. Res. 2021, 402, 108038. [Google Scholar] [CrossRef]
  5. Shen, Y.; Ye, B.; Chen, P.; Wang, Q.; Fan, C.; Shu, Y.; Xiang, M. Cognitive decline, dementia, Alzheimer’s disease and presbycusis: Examination of the possible molecular mechanism. Front. Neurosci. 2018, 12, 394. [Google Scholar] [CrossRef]
  6. Pizzino, G.; Irrera, N.; Cucinotta, M.; Pallio, G.; Mannino, F.; Arcoraci, V.; Bitto, A. Oxidative stress: Harms and benefits for human health. Oxid. Med. Cell. Longev. 2017, 2017, 8416763. [Google Scholar] [CrossRef] [Green Version]
  7. Guo, C.; Sun, L.; Chen, X.; Zhang, D. Oxidative stress, mitochondrial damage and neurodegenerative diseases. Neural Regen. Res. 2013, 8, 2003. [Google Scholar] [CrossRef]
  8. Van Houten, B.; Woshner, V.; Santos, J.H. Role of mitochondrial DNA in toxic responses to oxidative stress. DNA Repair 2006, 5, 145–152. [Google Scholar] [CrossRef]
  9. Mao, P. Oxidative stress and its clinical applications in dementia. J. Neurodegener. Dis. 2013, 2013, 319898. [Google Scholar] [CrossRef] [Green Version]
  10. Girard, P.M.; Boiteux, S. Repair of oxidized DNA bases in the yeast Saccharomyces cerevisiae. Biochimie 1997, 79, 559–566. [Google Scholar] [CrossRef]
  11. Durany, N.; Münch, G.; Michel, T.; Riederer, P. Investigations on oxidative stress and therapeutical implications in dementia. Eur. Arch. Psychiatry Clin. Neurosci. 1999, 249, S68–S73. [Google Scholar] [CrossRef]
  12. Christen, Y. Oxidative stress and Alzheimer disease. Am. J. Clin. Nutr. 2000, 71, 621S–629S. [Google Scholar] [CrossRef] [Green Version]
  13. Fujimoto, C.; Yamasoba, T. Oxidative stresses and mitochondrial dysfunction in age-related hearing loss. Oxidative Med. Cell. Longev. 2014, 2014, 582849. [Google Scholar] [CrossRef] [Green Version]
  14. Yamasoba, T.; Lin, F.R.; Someya, S.; Kashio, A.; Sakamoto, T.; Kondo, K. Current concepts in age-related hearing loss: Epidemiology and mechanistic pathways. Hear. Res. 2013, 303, 30–38. [Google Scholar] [CrossRef] [Green Version]
  15. Bai, U.; Seidman, M.D.; Hinojosa, R.; Quirk, W.S. Mitochondrial DNA deletions associated with aging and possibly presbycusis: A human archival temporal bone study. Otol. Neurotol. 1997, 18, 449–453. [Google Scholar]
  16. Karimian, M.; Behjati, M.; Barati, E.; Ehteram, T.; Karimian, A. CYP1A1 and GST s common gene variations and presbycusis risk: A genetic association analysis and a bioinformatics approach. Environ. Sci. Pollut. Res. 2020, 27, 42600–42610. [Google Scholar] [CrossRef] [PubMed]
  17. Bared, A.; Ouyang, X.; Angeli, S.; Du, L.L.; Hoang, K.; Yan, D.; Liu, X.Z. Antioxidant enzymes, presbycusis, and ethnic variability. Otolaryngol. Head Neck Surg. 2010, 143, 263–268. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  18. Manche, S.K.; Jangala, M.; Putta, P.; Koralla, R.M.; Akka, J. Association of oxidative stress gene polymorphisms with presbycusis. Gene 2016, 593, 277–283. [Google Scholar] [CrossRef]
  19. Bloem, B.R.; Okun, M.S.; Klein, C. Parkinson’s disease. Lancet 2021, 397, 2284–2303. [Google Scholar] [CrossRef]
  20. McGeer, P.L.; McGeer, E.G. Glial reactions in Parkinson’s disease. Mov. Disord. 2008, 23, 474–483. [Google Scholar] [CrossRef]
  21. Madetko, N.; Migda, B.; Alster, P.; Turski, P.; Koziorowski, D.; Friedman, A. Platelet-to-lymphocyte ratio and neutrophil-tolymphocyte ratio may reflect differences in PD and MSA-P neuroinflammation patterns. Neurol. Neurochir. Pol. 2022, 56, 148–155. [Google Scholar] [CrossRef] [PubMed]
  22. Chang, K.H.; Chen, C.M. The role of oxidative stress in Parkinson’s disease. Antioxidants 2020, 9, 597. [Google Scholar] [CrossRef]
  23. Langston, J.W.; Ballard, P.; Tetrud, J.W.; Irwin, I. Chronic Parkinsonism in humans due to a product of meperidine-analog synthesis. Science 1983, 219, 979–980. [Google Scholar] [CrossRef] [Green Version]
  24. Wei, Z.; Li, X.; Li, X.; Liu, Q.; Cheng, Y. Oxidative stress in Parkinson’s disease: A systematic review and meta-analysis. Front. Mol. Neurosci. 2018, 11, 236. [Google Scholar] [CrossRef] [PubMed]
  25. Jiménez-Jiménez, F.J.; Alonso-Navarro, H.; García-Martín, E.; Agúndez, J.A. Coenzyme Q10 and Parkinsonian Syndromes: A Systematic Review. J. Pers. Med. 2022, 12, 975. [Google Scholar] [CrossRef] [PubMed]
  26. Uversky, V.N. A protein-chameleon: Conformational plasticity of α-synuclein, a disordered protein involved in neurodegenerative disorders. J. Biomol. Struct. Dyn. 2003, 21, 211–234. [Google Scholar] [CrossRef]
  27. Bonini, N.M.; Giasson, B.I. Snaring the function of α-synuclein. Cell 2005, 123, 359–361. [Google Scholar] [CrossRef]
  28. Hamilton, R.L. Lewy bodies in Alzheimer’s disease: A neuropathological review of 145 cases using α-synuclein immunohistochemistry. Brain Pathol. 2000, 10, 378–384. [Google Scholar] [CrossRef]
  29. Arai, Y.; Yamazaki, M.; Mori, O.; Muramatsu, H.; Asano, G.; Katayama, Y. α-Synuclein-positive structures in cases with sporadic Alzheimer’s disease: Morphology and its relationship to tau aggregation. Brain Res. 2001, 888, 287–296. [Google Scholar] [CrossRef]
  30. Lippa, C.F.; Fujiwara, H.; Mann, D.M.; Giasson, B.; Baba, M.; Schmidt, M.L.; Nee, L.E.; O’Connell, B.; Pollen, D.A.; George-Hyslop, P.S.; et al. Lewy bodies contain altered α-synuclein in brains of many familial Alzheimer’s disease patients with mutations in presenilin and amyloid precursor protein genes. Am. J. Pathol. 1998, 153, 1365–1370. [Google Scholar] [CrossRef]
  31. Larson, M.E.; Sherman, M.A.; Greimel, S.; Kuskowski, M.; Schneider, J.A.; Bennett, D.A.; Lesné, S.E. Soluble α-synuclein is a novel modulator of Alzheimer’s disease pathophysiology. J. Neurosci. 2012, 32, 10253–10266. [Google Scholar] [CrossRef] [Green Version]
  32. Twohig, D.; Rodriguez-Vieitez, E.; Sando, S.B.; Berge, G.; Lauridsen, C.; Møller, I.; Grøntvedt, G.R.; Bråthen, G.; Patra, K.; Bu, G.; et al. The relevance of cerebrospinal fluid α-synuclein levels to sporadic and familial Alzheimer’s disease. Acta Neuropathol. Commun. 2018, 6, 130. [Google Scholar] [CrossRef] [Green Version]
  33. Korff, A.; Liu, C.; Ginghina, C.; Shi, M.; Zhang, J. Alzheimer’s Disease Neuroimaging Initiative—α-Synuclein in cerebrospinal fluid of Alzheimer’s disease and mild cognitive impairment. J. Alzheimers Dis. 2013, 36, 679–688. [Google Scholar] [CrossRef]
  34. Park, S.N.; Back, S.A.; Choung, Y.H.; Kim, H.L.; Akil, O.; Lustig, L.R.; Park, K.H.; Yeo, S.W. alpha-Synuclein deficiency and efferent nerve degeneration in the mouse cochlea: A possible cause of early-onset presbycusis. Neurosci. Res. 2011, 71, 303–310. [Google Scholar] [CrossRef]
  35. Akil, O.; Chang, J.; Pak, V.; Habib, M.; Lustig, L. Alpha-synuclein Overexpression Causes Mild Hearing Loss and Protection from Noise Trauma. Otolaryngol. Open Access J. (OOAJ) 2022, 7. [Google Scholar] [CrossRef]
  36. Belin, A.C.; Westerlund, M. Parkinson’s disease: A genetic perspective. FEBS J. 2008, 275, 1377–1383. [Google Scholar] [CrossRef]
  37. Campêlo, C.L.; Silva, R.H. Genetic variants in SNCA and the risk of sporadic Parkinson’s disease and clinical outcomes: A review. Park. Dis. 2017, 2017, 4318416. [Google Scholar] [CrossRef] [Green Version]
  38. Goedert, M.; Spillantini, M.G.; Del Tredici, K.; Braak, H. 100 years of Lewy pathology. Nat. Rev. Neurol. 2013, 9, 13–24. [Google Scholar] [CrossRef] [PubMed]
  39. Scudamore, O.; Ciossek, T. Increased oxidative stress exacerbates α-synuclein aggregation in vivo. J. Neuropathol. Exp. Neurol. 2018, 77, 443–453. [Google Scholar] [CrossRef] [PubMed]
  40. Hashimoto, M.; Hsu, L.J.; Xia, Y.; Takeda, A.; Sisk, A.; Sundsmo, M.; Masliah, E. Oxidative stress induces amyloid-like aggregate formation of NACP/α-synuclein in vitro. Neuroreport 1999, 10, 717–721. [Google Scholar] [CrossRef] [PubMed]
  41. Parihar, M.S.; Parihar, A.; Fujita, M.; Hashimoto, M.; Ghafourifar, P. Alpha-synuclein overexpression and aggregation exacerbates impairment of mitochondrial functions by augmenting oxidative stress in human neuroblastoma cells. Int. J. Biochem. Cell Biol. 2009, 41, 2015–2024. [Google Scholar] [CrossRef]
  42. Lotharius, J.; Brundin, P. Pathogenesis of Parkinson’s disease: Dopamine, vesicles and α-synuclein. Nat. Rev. Neurosci. 2002, 3, 932–942. [Google Scholar] [CrossRef] [PubMed]
  43. Bhatia, K.P.; Bain, P.; Bajaj, N.; Elble, R.J.; Hallett, M.; Louis, E.D.; Raethjen, J.; Stamelou, M.; Testa, C.M.; Deuschl, G. Tremor Task Force of the International Parkinson and Movement Disorder Society—Consensus Statement on the classification of tremors: From the task force on tremor of the International Parkinson and Movement Disorder Society. Mov. Disord. 2018, 33, 75–87. [Google Scholar] [CrossRef] [PubMed]
  44. Tarakad, A.; Jankovic, J. Essential tremor and Parkinson’s disease: Exploring the relationship. Tremor Other Hyperkin. Mov. 2018, 8, 589. [Google Scholar] [CrossRef]
  45. Benito-Leon, J.; Louis, E.D.; Bermejo-Pareja, F. Risk of incident Parkinson’s disease and parkinsonism in essential tremor: A population based study. J. Neurol. Neurosurg. Psychiatry 2009, 80, 423–425. [Google Scholar] [CrossRef] [PubMed]
  46. Louis, E.D. Non-motor symptoms in essential tremor: A review of the current data and state of the field. Park. Relat. Disord. 2016, 22, S115–S118. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  47. Ondo, W.G.; Sutton, L.; Vuong, K.D.; Lai, D.; Jankovic, J. Hearing impairment in essential tremor. Neurology 2003, 61, 1093–1097. [Google Scholar] [CrossRef]
  48. Balaban, H.; Altuntaş, E.E.; Uysal, İ. Ö; Şentürk, İ.A.; Topaktaş, S. Audio-vestibular evaluation in patients with essential tremor. Eur. Arch. Oto-Rhino-Laryngol. 2012, 269, 1577–1581. [Google Scholar] [CrossRef]
  49. Leme, M.S.; Sanches, S.G.; Carvallo, R.M. Peripheral hearing in Parkinson’s disease: A systematic review. Int. J. Audiol. 2022. [Google Scholar] [CrossRef]
  50. Shetty, K.; Krishnan, S.; Thulaseedharan, J.V.; Mohan, M.; Kishore, A. Asymptomatic hearing impairment frequently occurs in early-onset Parkinson’s disease. J. Mov. Disord. 2019, 12, 84. [Google Scholar] [CrossRef]
  51. Pisani, V.; Sisto, R.; Moleti, A.; Di Mauro, R.; Pisani, A.; Brusa, L.; Altavista, M.C.; Stanzione, P.; Di Girolamo, S. An investigation of hearing impairment in de novo Parkinson’s disease patients: A preliminary study. Park. Relat. Disord. 2015, 21, 987–991. [Google Scholar] [CrossRef]
  52. Jafari, Z.; Kolb, B.E.; Mohajerani, M.H. Auditory dysfunction in Parkinson’s disease. Mov. Disord. 2020, 35, 537–550. [Google Scholar] [CrossRef]
  53. Schrag, A.; Bohlken, J.; Dammertz, L.; Teipel, S.; Hermann, W.; Akmatov, M.K.; Bätzing, J.; Holstiege, J. Widening the spectrum of risk factors, comorbidities, and prodromal features of Parkinson disease. JAMA Neurol. 2023, 80, 161–171. [Google Scholar] [CrossRef] [PubMed]
  54. Lai, S.W.; Liao, K.F.; Lin, C.L.; Lin, C.C.; Sung, F.C. Hearing loss may be a non-motor feature of Parkinson’s disease in older people in Taiwan. Eur. J. Neurol. 2014, 21, 752–757. [Google Scholar] [CrossRef] [PubMed]
  55. Simonet, C.; Bestwick, J.; Jitlal, M.; Waters, S.; Ben-Joseph, A.; Marshall, C.R.; Dobson, R.; Marrium, S.; Robson, J.; Jacobs, B.M.; et al. Assessment of risk factors and early presentations of Parkinson disease in primary care in a diverse UK population. JAMA Neurol. 2022, 79, 359–369. [Google Scholar] [CrossRef] [PubMed]
  56. Steptoe, A.; Breeze, E.; Banks, J.; Nazroo, J. Cohort profile: The English longitudinal study of ageing. Int. J. Epidemiol. 2013, 42, 1640–1648. [Google Scholar] [CrossRef] [Green Version]
  57. Davis, A.; Smith, P.; Ferguson, M.; Stephens, D.; Gianopoulos, I. Acceptability, benefit and costs of early screening for hearing disability: A study of potential screening tests and models. Health Technol. Assess 2007, 11, 1–294. [Google Scholar] [CrossRef]
  58. Scholes, S.; Biddulph, J.; Davis, A.; Mindell, J.S. Socioeconomic differences in hearing among middle-aged and older adults: Cross-sectional analyses using the health survey for England. BMJ Open 2018, 8, e019615. [Google Scholar] [CrossRef] [Green Version]
  59. Reeve, A.; Simcox, E.; Turnbull, D. Ageing and Parkinson’s disease: Why is advancing age the biggest risk factor? Ageing Res. Rev. 2014, 14, 19–30. [Google Scholar] [CrossRef]
  60. Deafness and Hearing Loss. Available online: https://www.who.int/news-room/fact-sheets/detail/deafness-and-hearing-loss (accessed on 17 February 2023).
  61. Henderson, R.; Diggle, P.; Dobson, A. Joint modelling of longitudinal measurements and event time data. Biostatistics 2000, 1, 465–480. [Google Scholar] [CrossRef]
  62. Crowther, M.J.; Abrams, K.R.; Lambert, P.C. Joint modeling of longitudinal and survival data. Stata J. 2013, 13, 165–184. [Google Scholar] [CrossRef] [Green Version]
  63. Kim, S.Y.; Kim, H.J.; Kim, M.S.; Park, B.; Kim, J.H.; Choi, H.G. Discrepancy between self-assessed hearing status and measured audiometric evaluation. PLoS ONE 2017, 12, e0182718. [Google Scholar] [CrossRef] [Green Version]
  64. Choi, J.E.; Moon, I.J.; Baek, S.Y.; Kim, S.W.; Cho, Y.S. Discrepancies between self-reported hearing difficulty and hearing loss diagnosed by audiometry: Prevalence and associated factors in a national survey. BMJ Open 2019, 9, e022440. [Google Scholar] [CrossRef] [Green Version]
  65. Tsimpida, D.; Kontopantelis, E.; Ashcroft, D.; Panagioti, M. Comparison of self-reported measures of hearing with an objective audiometric measure in adults in the English longitudinal study of ageing. JAMA Netw. Open 2020, 3, e2015009. [Google Scholar] [CrossRef]
  66. Fortnum, H.; Ukoumunne, O.C.; Hyde, C.; Taylor, R.S.; Ozolins, M.; Errington, S.; Zhelev, Z.; Pritchard, C.; Benton, C.; Moody, J.; et al. Diagnostic accuracy of the pure-tone screen and HearCheck screener for identifying hearing impairment in school children. In A Programme of Studies Including Assessment of Diagnostic Accuracy of School Hearing Screening Tests and a Cost-Effectiveness Model of School Entry Hearing Screening Programmes; NIHR Journals Library: London, UK, 2016. [Google Scholar]
  67. Keithley, E.M. Pathology and mechanisms of cochlear aging. J. Neurosci. Res. 2020, 98, 1674–1684. [Google Scholar] [CrossRef] [Green Version]
  68. Updated Guidance on Onward Referral. Available online: https://www.bshaa.org/updated-guidance-on-onward-referral/ (accessed on 12 May 2023).
  69. Yang, T.H.; Chu, Y.C.; Chen, Y.F.; Chen, M.Y.; Cheng, Y.F.; Wu, C.S.; Huang, H.M. Diagnostic validity of self-reported hearing loss in elderly taiwanese individuals: Diagnostic performance of a hearing self-assessment questionnaire on audiometry. Int. J. Environ. Res. Public Health 2021, 18, 13215. [Google Scholar] [CrossRef]
  70. Pickles, J.O. An Introduction to the Physiology of Hearing, 4th ed.; Emerald Publishing: Bingley, UK, 2013. [Google Scholar]
  71. Carl, A.; Hohman, M.; Cornejo, J. Audiology Pure Tone Evaluation; StatPearls: St. Petersburg, FL, USA, 2023. [Google Scholar] [CrossRef]
  72. Wallhagen, M.I. The stigma of hearing loss. Gerontologist 2010, 50, 66–75. [Google Scholar] [CrossRef]
  73. Jankovic, J. Parkinson’s disease: Clinical features and diagnosis. J. Neurol. Neurosurg. Psychiatry 2008, 79, 368–376. [Google Scholar] [CrossRef] [Green Version]
  74. Dickson, D.W. Parkinson’s disease and parkinsonism: Neuropathology. Cold Spring Harb. Perspect. Med. 2012, 2, a009258. [Google Scholar] [CrossRef] [Green Version]
  75. Vieira, B.D.; Radford, R.A.; Chung, R.S.; Guillemin, G.J.; Pountney, D.L. Neuroinflammation in multiple system atrophy: Response to and cause of α-synuclein aggregation. Front. Cell. Neurosci. 2015, 9, 437. [Google Scholar] [CrossRef]
  76. Monzio Compagnoni, G.; Di Fonzo, A. Understanding the pathogenesis of multiple system atrophy: State of the art and future perspectives. Acta Neuropathol. Commun. 2019, 7, 113. [Google Scholar] [CrossRef] [Green Version]
  77. Alster, P.; Madetko, N.; Koziorowski, D.; Friedman, A. Microglial activation and inflammation as a factor in the pathogenesis of progressive supranuclear palsy (PSP). Front. Neurosci. 2020, 14, 893. [Google Scholar] [CrossRef]
  78. Matsuoka, K.; Takado, Y.; Tagai, K.; Kubota, M.; Sano, Y.; Takahata, K.; Ono, M.; Seki, C.; Matsumoto, H.; Endo, H.; et al. Two pathways differentially linking tau depositions, oxidative stress, and neuronal loss to apathetic phenotypes in progressive supranuclear palsy. J. Neurol. Sci. 2023, 444, 120514. [Google Scholar] [CrossRef]
  79. Ali, F.; Josephs, K.A. Corticobasal degeneration: Key emerging issues. J. Neurol. 2018, 265, 439–445. [Google Scholar] [CrossRef]
  80. Castellani, R.; Smith, M.A.; Richey, P.L.; Kalaria, R.; Gambetti, P.; Perry, G. Evidence for oxidative stress in Pick disease and corticobasal degeneration. Brain Res. 1995, 696, 268–271. [Google Scholar] [CrossRef]
  81. Chai, Y.L.; Lee, J.H.; Chong, J.R.; Ballard, C.; Francis, P.T.; Kennedy, B.K.; Arumugam, T.V.; Chen, C.P.; Aarsl, D.; Lai, M.K. Inflammatory panel cytokines are elevated in the neocortex of late-stage Alzheimer’s disease but not Lewy body dementias. J. Neuroinflamm. 2023, 20, 111. [Google Scholar] [CrossRef]
  82. Navarro, A.; Boveris, A.; Bández, M.J.; Sánchez-Pino, M.J.; Gómez, C.; Muntané, G.; Ferrer, I. Human brain cortex: Mitochondrial oxidative damage and adaptive response in Parkinson disease and in dementia with Lewy bodies. Free Radic. Biol. Med. 2009, 46, 1574–1580. [Google Scholar] [CrossRef]
  83. Hughes, A.J.; Daniel, S.E.; Kilford, L.; Lees, A.J. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: A clinico-pathological study of 100 cases. J. Neurol. Neurosurg. Psychiatry 1992, 55, 181–184. [Google Scholar] [CrossRef] [Green Version]
  84. Nicoletti, A.; Luca, A.; Baschi, R.; Cicero, C.E.; Mostile, G.; Davì, M.; Pilati, L.; Restivo, V.; Zappia, M.; Monastero, R. Incidence of mild cognitive impairment and dementia in Parkinson’s disease: The Parkinson’s disease cognitive impairment study. Front. Aging Neurosci. 2019, 11, 21. [Google Scholar] [CrossRef] [Green Version]
  85. Irwin, D.J.; Lee, V.M.; Trojanowski, J.Q. Parkinson’s disease dementia: Convergence of α-synuclein, tau and amyloid-β pathologies. Nat. Rev. Neurosci. 2013, 14, 626–636. [Google Scholar] [CrossRef]
  86. Fereshtehnejad, S.M.; Zeighami, Y.; Dagher, A.; Postuma, R.B. Clinical criteria for subtyping Parkinson’s disease: Biomarkers and longitudinal progression. Brain 2017, 140, 1959–1976. [Google Scholar] [CrossRef]
  87. Movement Disorder Society Task Force on Rating Scales for Parkinson’s Disease. The unified Parkinson’s disease rating scale (UPDRS): Status and recommendations. Mov. Disord. 2003, 18, 738–750. [Google Scholar] [CrossRef]
  88. Powell, D.S.; Oh, E.S.; Reed, N.S.; Lin, F.R.; Deal, J.A. Hearing loss and cognition: What we know and where we need to go. Front. Aging Neurosci. 2022, 13, 769405. [Google Scholar] [CrossRef]
  89. Lin, F.R.; Albert, M. Hearing loss and dementia—Who is listening? Aging Ment. Health 2014, 18, 671–673. [Google Scholar] [CrossRef] [Green Version]
  90. Dawes, P.; Emsley, R.; Cruickshanks, K.J.; Moore, D.R.; Fortnum, H.; Edmondson-Jones, M.; McCormack, A.; Munro, K.J. Hearing loss and cognition: The role of hearing AIDS, social isolation and depression. PLoS ONE 2015, 10, e0119616. [Google Scholar] [CrossRef] [Green Version]
  91. Lin, F.R.; Pike, J.R.; Albert, M.S.; Arnold, M.; Burgard, S.; Chisolm, T.; Couper, D.; Deal, J.A.; Goman, A.M.; Glynn, N.W.; et al. Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): A multicentre, randomised controlled trial. Lancet 2023. [Google Scholar] [CrossRef]
  92. The Incidence and Prevalence of Parkinson’s in the UK. Available online: https://www.parkinsons.org.uk/professionals/resources/methods-used-estimate-incidence-and-prevalence-parkinsons-uk-report (accessed on 12 May 2023).
  93. The English Longitudinal Study of Ageing Technical Reports. Available online: https://www.elsa-project.ac.uk/wave-reports (accessed on 12 May 2023).
  94. Hernández, A.V.; Eijkemans, M.J.; Steyerberg, E.W. Randomized controlled trials with time-to-event outcomes: How much does prespecified covariate adjustment increase power? Ann. Epidemiol. 2006, 16, 41–48. [Google Scholar] [CrossRef]
  95. Montesinos López, O.A.; Montesinos López, A.; Crossa, J. Overfitting, model tuning, and evaluation of prediction performance. In Multivariate Statistical Machine Learning Methods for Genomic Prediction; Springer International Publishing: Cham, Switzerland, 2022; pp. 109–139. [Google Scholar]
  96. Babyak, M.A. What you see may not be what you get: A brief, nontechnical introduction to overfitting in regression-type models. Psychosom. Med. 2004, 66, 411–421. [Google Scholar]
  97. Bates, D.; Kliegl, R.; Vasishth, S.; Baayen, H. Parsimonious mixed models. arXiv 2015, arXiv:1506.04967. [Google Scholar] [CrossRef]
Table 1. Joint longitudinal and survival model.
Table 1. Joint longitudinal and survival model.
MeanSDConfidence Intervalp
Longitudinal Outcome
Intercept1.3570.1131.135–1.581<0.001
Cohort Y => 21.8120.0191.774–1.847<0.001
Cohort Y => 34.1790.0264.126–4.228<0.001
Cohort Y => 46.3380.0406.263–6.417<0.001
Age−0.0580.002−0.062–−0.055<0.001
Wave Number−0.0580.002−0.062–−0.053<0.001
Survival Outcome
Age0.540.0130.026–0.078<0.001
Hearing Loss0.0630.075−0.080–−0.2190.402
Table 2. Baseline demographic characteristics of study cohort by hearing loss status.
Table 2. Baseline demographic characteristics of study cohort by hearing loss status.
Hearing Loss Status
Normal (n = 3575)Moderate (n = 850)Moderately Severe/Severe (n = 387)p *
Sex, Male1416417232F(2,4809) = 459.9, p < 0.001
Age, mean (SE), years66.48 (0.283)72.62 (0.396)75.95 (0.454)F(1,4809) = 38.01, p < 0.001
Incident PD (n)1555F(2,4809)= 2.937, p = 0.053
* Note. The p-values documented here were obtained from between-subject ANOVAs which examined whether demographic (age and sex) and PD incidence significantly differ across hearing loss groups. In this analysis, hearing loss level formed the independent variable and the relevant demographic variable formed the dependent variable.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Readman, M.R.; Wan, F.; Fairman, I.; Linkenauger, S.A.; Crawford, T.J.; Plack, C.J. Is Hearing Loss a Risk Factor for Idiopathic Parkinson’s Disease? An English Longitudinal Study of Ageing Analysis. Brain Sci. 2023, 13, 1196. https://doi.org/10.3390/brainsci13081196

AMA Style

Readman MR, Wan F, Fairman I, Linkenauger SA, Crawford TJ, Plack CJ. Is Hearing Loss a Risk Factor for Idiopathic Parkinson’s Disease? An English Longitudinal Study of Ageing Analysis. Brain Sciences. 2023; 13(8):1196. https://doi.org/10.3390/brainsci13081196

Chicago/Turabian Style

Readman, Megan Rose, Fang Wan, Ian Fairman, Sally A. Linkenauger, Trevor J. Crawford, and Christopher J. Plack. 2023. "Is Hearing Loss a Risk Factor for Idiopathic Parkinson’s Disease? An English Longitudinal Study of Ageing Analysis" Brain Sciences 13, no. 8: 1196. https://doi.org/10.3390/brainsci13081196

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop