Hearing Loss and Depression in Older Adults

A Systematic Review and Meta-analysis

Blake J. Lawrence, PhD; Dona M. P. Jayakody, PhD; Rebecca J. Bennett, PhD; Robert H. Eikelboom, PhD; Natalie Gasson, PhD; Peter L. Friedland, MBBCh, FRACS

Disclosures

Gerontologist. 2020;60(3):e137-e154. 

In This Article

Abstract and Introduction

Abstract

Background and Objectives: Studies reporting an association between hearing loss and depression in older adults are conflicting and warrant a systematic review and meta-analysis of the evidence.

Research Design and Methods: A search of academic databases (e.g., MEDLINE) and gray literature (e.g., OpenGrey) identified relevant articles published up to July 17, 2018. Cross-sectional or cohort designs were included. Outcome effects were computed as odds ratios (ORs) and pooled using random-effects meta-analysis (PROSPERO: CRD42018084494).

Results: A total of 147,148 participants from 35 studies met inclusion criteria. Twenty-four studies were cross-sectional and 11 were cohort designs. Overall, hearing loss was associated with statistically significantly greater odds of depression in older adults (OR = 1.47, 95% confidence interval [CI] = 1.31–1.65). When studies were stratified by design, hearing loss was associated with greater odds of depression in cross-sectional studies (OR = 1.54, 95% CI = 1.31–1.80) and cohort studies (OR = 1.39, 95% CI = 1.16 – 1.67), and there was no difference between cross-sectional or cohort effect estimates (Q = 0.64, p = .42). There was no effect of moderator variables (i.e., hearing aid use) on the association between hearing loss and depression, but these findings must be interpreted with caution. There was no presence of publication bias but certainty in the estimation of the overall effect was classified as "low."

Discussion and Implications: Older adults may experience increased odds of depression associated with hearing loss, and this association may not be influenced by study or participant characteristics.

Introduction

Global estimates indicate that over 1.30 billion people currently live with some form of hearing loss and this prevalence will likely rise with the aging population (Wilson, Tucci, Merson, & O'Donoghue, 2017). For older adults (≥60 years of age), hearing loss is often caused by the loss of inner and outer hair cells at the basal end of the basilar membrane, which contributes to the loss of high frequency hearing and increased hearing thresholds experienced during aging (Peelle & Wingfield, 2016). Approximately 13% of adults 40–49 years of age experience some form of hearing loss, whereas almost 45% of older adults aged 60–69 years live with hearing loss and this prevalence increases to 90% for adults 80 years and older (Goman & Lin, 2016). Aging may also be associated with increased risk of depression (Freeman et al., 2016), which is characterized by sadness, feelings of low self-worth or guilt, a loss of interest in daily activities, and disturbed appetite or sleep, which affect concentration (World Health Organization, 2018). Approximately 15% of older adults experience mild depressive symptoms and 1%–5% live with major depressive disorder (Fiske, Wetherell, & Gatz, 2009). Moreover, research has shown an association between hearing loss and depression in older adults (e.g., Keidser & Seeto, 2017; Rosso et al., 2013), with age-related changes in psychosocial experience as well as degeneration to cortical activity proposed to explain these concomitant conditions.

The association between hearing loss and depression in older adults has most frequently been examined within, and explained by, the potential influence of psychosocial changes experienced during aging. Kiely, Anstey, and Luszcz (2013) initially found that severity of depressive symptoms was associated with hearing loss in older adults, but this association reduced to nonsignificance (i.e., was accounted for) when difficulty completing daily activities and degree of social engagement in daily life were included in the model. Further evidence shows a faster decline in hearing is associated with greater social and emotional loneliness among older adults (Pronk et al., 2014), and when left untreated, hearing loss may develop into a chronic stressor that leads to the proliferation of depression as an additional stressor (West, 2017). Hearing loss may therefore worsen existing difficulties associated with psychosocial and functional abilities during older age, increasing the likelihood of developing depression. Conversely, recent evidence shows no influence of psychosocial factors (i.e., less participation in social activities or access to a social network) on the association between hearing loss and depression in older adults (Cosh et al., 2018). The authors proposed that older adults may accept hearing loss as part of the normal aging experience and therefore adapt to changes in their hearing by modifying/improving their communication skills or using hearing aids to alleviate the burden of hearing loss, which in turn mitigates the potential negative impact of hearing loss on psychosocial experiences that may precede depression (Cosh et al., 2018). It is therefore not clear if the association between hearing loss and depression in older adults can be explained by a psychosocial mechanism in later life. A meta-analysis and systematic review of the literature will provide an estimate of this comorbid association and the potential influence of psychosocial or health characteristics that may account for this relationship in older adults.

Recent evidence also suggests that degeneration to neuropathological mechanisms associated with auditory perception and regulation of mood may explain the relationship between hearing loss and depression in older adults. As extensively reviewed by Rutherford, Brewster, Golub, Kim, and Roose (2018), neuroimaging studies show similar patterns of diminished activity in the limbic system (responsible for emotion and behavior), the frontal cortex (responsible for emotional regulation, reasoning, and planning), and auditory cortex in older adults with hearing loss or depression. These initial findings suggest the presence of common neural degeneration associated with hearing loss and depression in older adults, but more evidence is needed to increase our understanding of the pathophysiology underlying hearing loss and depression in later life.

To date, the association between hearing loss and depression is most frequently reported in epidemiological studies. Some cross-sectional studies report an association between hearing loss and depression during older age (Behera et al., 2016; Keidser & Seeto, 2017; Lee & Hong, 2016), whereas others report no relationship (Bergdahl et al., 2005; Chou & Chi, 2005). Similar evidence exists for cohort studies, with initial results showing hearing loss is associated with increased odds of depression among older adults (Forsell, 2000) and later studies repudiating these findings (Chou, 2008; (Cosh et al., 2018). Conflicting findings in the literature may be the consequence of methodological variance between studies and the limitations associated with epidemiological research. As a method of investigation, epidemiology allows for the examination of health characteristics within large population-based samples of participants when conducting a more controlled clinical trial is not feasible (e.g., examining intercontinental dietary patterns) or ethical (e.g., examining the health effects of smoking). However, epidemiological studies are often influenced by biases that undermine reliability in their results. As reported by Ioannidis (2016), most initial statistically significant epidemiological findings are later not replicated in more scientifically robust randomized controlled trials. Moreover, large longitudinal epidemiological studies examining changes in health characteristics (e.g., U.S. National Health and Nutrition Epidemiological Survey) often find statistically significant correlations between almost all variables of interest (Patel, Ioannidis, Cullen, & Rehkopf, 2015). With these limitations in mind, however, a meta-analysis and systematic review of epidemiological (e.g., cross-sectional and cohort) studies can provide a more rigorous estimate of an association between health characteristics (e.g., hearing loss and depression), while drawing attention to the strengths and weaknesses within the existing evidence and providing recommendations for future clinical practice.

Previous studies investigating the association between hearing loss and depression have estimated hearing loss using objective measures such as pure tone audiometry (Hidalgo et al., 2009; Kiely et al., 2013), but a proportion of studies only report subjective hearing loss measured by self-report outcomes (e.g., Boorsma et al., 2012; Saito et al., 2010). Some of these studies included a proportion of participants with cognitive decline (Perlmutter, Bhorade, Gordon, Hollingsworth, & Baum, 2010). Described as deficits in cognitive function not normal for age and level of education, cognitive decline in older adults may range from mild cognitive impairment to dementia (Albert et al., 2011). Increasing evidence shows an association between hearing loss and cognitive decline in older adults (Loughrey, Kelly, Kelley, Brennan, & Lawlor, 2017), and a decline in cognitive function has been associated with depression (Wang & Blazer, 2015). We may therefore expect a stronger association between hearing loss and depression in older adults that also demonstrate presence of cognitive decline (Rutherford et al., 2018). Participant experience using hearing aids also varies across studies (Chou, 2008; Pronk et al., 2011; Rosso et al., 2013). Hearing aids may alleviate depressive symptoms associated with hearing loss in older adults (Choi et al., 2016; Manrique-Huarte, Calavia, Irujo, Girón, & Manrique-Rodríguez, 2016), which may influence the association between hearing loss and depression in observational research. In addition, a high proportion of studies do not report outcome results adjusted for the confounding influence of covariates (e.g., health/psychosocial characteristics), which undermines validity of their findings (Al Sabahi, Al Sinawi, Al Hinai, & Youssef, 2014; Chou & Chi, 2005; Hidalgo et al., 2009). It is also well-understood that cross-sectional studies cannot infer causality and are therefore unable to determine if there is temporal relationship between hearing loss and depression. Initial meta-analytic evidence from a small number of studies showed an association between hearing loss and depression in later life (Huang, Dong, Lu, Yue, & Liu, 2010), but more studies have since been published and inconsistency in the findings warrants the need for a systematic review and meta-analysis of the evidence.

The primary aim of this systematic review and meta-analysis was to synthesize the available evidence to provide a summary effect estimate of the association between hearing loss and depression in older adults. The secondary aim was to examine whether study (e.g., design, outcome measures) or participant (e.g., demographic, health) characteristics may influence the association between hearing loss and depression. An extensive systematic review of the literature was conducted and all available evidence was included in this study to provide the most rigorous estimate of the association between hearing loss and depression in older adults.

processing....