Introduction: Hearing loss (HL) is considered a potentially modifiable risk factor for dementia. We aimed to examine the relationship between HL and incident dementia diagnosis in a province-wide population-based cohort study with matched controls. Methods: Administrative healthcare databases were linked to generate a cohort of patients who were aged ≥40 years at their first claimed hearing amplification devices (HAD) between April 2007 and March 2016 through the Assistive Devices Program (ADP) (257,285 with claims and 1,005,010 controls). The main outcome was incident dementia diagnosis, ascertained using validated algorithms. Dementia incidence was compared between cases and controls using Cox regression. Patient, disease, and other risk factors were examined. Results: Dementia incidence rates (per 1,000 person-years) were 19.51 (95% confidence interval [CI]: 19.26–19.77) and 14.15 (95% CI: 14.04–14.26) for the ADP claimants and matched controls, respectively. In adjusted analyses, risk of dementia was higher in ADP claimants compared with controls (hazard ratio [HR]: 1.10 [95% CI: 1.09–1.12, p < 0.001]). Subgroup analyses showed a dose-response gradient, with risk of dementia higher among patients with bilateral HADs (HR: 1.12 [95% CI: 1.10–1.14, p < 0.001]), and an exposure-response gradient, with increasing risk over time from April 2007-March 2010 (HR: 1.03 [95% CI: 1.01–1.06, p = 0.014]), April 2010-March 2013 (HR: 1.12 [95% CI: 1.09–1.15, p < 0.001]), and April 2013-March 2016 (HR: 1.19 [95% CI: 1.16–1.23, p < 0.001]). Conclusion: In this population-based study, adults with HL had an increased risk of being diagnosed with dementia. Given the implications of HL on dementia risk, understanding the effect of hearing interventions merits further investigation.

Hearing loss (HL) is a public health concern as an estimated 4.6 million (19%) Canadians aged 20–79 experience HL, which affects their ability to hear everyday speech [1]. The prevalence of HL is expected to increase due to a growing older population, which raises concerns given the known associations between HL and social isolation, depression, falls, and cognitive decline [2, 3]. Dementia is a syndrome that describes a group of neurocognitive disorders characterized by progressive deterioration in one or more cognitive domains (which may include memory, language expression and/or comprehension, reasoning, and/or spatiotemporal perception), resulting in impairments in daily functioning and possible alterations in behavior [4]. It is estimated that the prevalence of dementia in Canada will double in the next 20 years; therefore, identifying modifiable risk factors is essential [5].

Despite the burden of dementia, their causes remain unknown [6, 8]. Epidemiological studies have shown that HL correlates with cognitive impairment and may be associated with a higher risk of incident dementia [9, 11]. Recently, HL has been identified as the top modifiable risk factor for dementia, accounting for 9% of risk reduction [12]. The possible mechanisms for the relationship between HL and dementia include the common pathology affecting the auditory system, impoverished environment, and increased cognitive resources needed for listening [13, 15]. A recent meta-analysis demonstrated that the use of hearing amplification devices (HADs) by patients with HL was associated with a 19% decrease in hazards of long-term cognitive decline [16]. As such, managing HL with HADs may be a preventative strategy for dementia [17]. Although previous studies have implicated HL as a risk factor for incident dementia, these findings require further confirmation through novel approaches to demonstrate reproducibility and generalizability. The objective of this study is to leverage health databases from a single-payer healthcare system to identify a large population-based cohort of adults aged 40 and older with HL and compare dementia diagnosis incidence with those of a matched population of controls.

Data Sources

Data utilized in this study were obtained from administrative datasets housed at the Institute of Clinical Evaluative Sciences (ICES), an independent, nonprofit research institute whose legal status under Ontario’s health information privacy law allows it to collect and analyze healthcare and demographic data, without consent, for health system evaluation. All personal identifying information is removed at ICES, and anonymous unique identifiers are generated for each patient.

The Assistive Devices Program (ADP) database contains information for claims of personalized assistive devices, including HADs. The Ontario Health Insurance Plan (OHIP) database contains information for fee-for-service claims submitted by physicians. Demographic information, including date of birth, sex, address, and date of death, was obtained with the Registered Persons Database (RPDB), a repository for all Ontario residents who are eligible for OHIP. Chronic comorbid conditions, including asthma, hypertension, diabetes, and dementia, were obtained with ICES-derived databases based on validated algorithms with linkage of the CIHI, OHIP, and RPDB datasets. These datasets were linked using unique encoded identifiers and analyzed at ICES.

Ontario is the largest province in Canada, with over 14 million residents in 2019, representing 38.8% of the Canadian population [18]. The Canadian healthcare system is publicly funded and provides universal coverage for all medically necessary services. The ADP allows Ontarians with valid OHIP cards to claim HADs and covers $500 CAD per ear every 5 years [19].

This study included 257,285 patients aged 40 and older from April 1, 2007, to March 31, 2016, who obtained HADs (unilateral/bilateral). HAD claims through ADP were used as a surrogate for creating a cohort of patients with HL. The study dates were selected to allow a follow-up of at least 5 years for identifying dementia diagnoses. Given the lack of audiometric data, the severity of HL was assessed in a subgroup analysis by comparing ADP claims for unilateral versus bilateral HADs. The study period was split into three segments (April 1, 2007–March 31, 2010; April 1, 2010–March 31, 2013; April 1, 2013–March 31, 2016) to analyze differences between early HADs exposure and outcomes. For each patient, corresponding controls were identified from the general population in RPDB. Using the ADP claim date as the index date for covariates, ADP claimants were matched to controls at a 1:5 ratio based on birth year and month (±1 month) and sex. We identified 1,005,010 matched controls. Exclusion criteria included patients with a prior dementia diagnosis, those aged 39 years or younger, and those with lapses in OHIP coverage or missing covariates of interest at the index date.

The primary outcome was incident dementia during the 5-year follow-up from April 1, 2007, to March 31, 2021, based on the following ICES-derived algorithm: at least one hospitalization record with a dementia diagnosis; at least three physician visits with a diagnosis of dementia on the claims that were 30 days apart in a 2-year period; or one prescription reimbursement record for any cholinesterase inhibitor. This algorithm has been validated using Ontario’s medical records [20].

We identified confounding covariates for the association between HL and diagnosis of dementia [7]. Sociodemographic information, including age, sex, geographical location, and neighborhood income quintiles, were obtained with RPDB. Comorbidities, including asthma, chronic obstructive pulmonary disease, congestive heart failure, stroke, acute myocardial infarction, and diabetes, were obtained.

Analysis

Summary statistics and weighted standardized differences between ADP claimants and the controls were calculated for socio-demographic characteristics and comorbidities. The Kaplan-Meier method described the time to the outcome of dementia diagnosis between ADP claimants and controls. Cox proportional hazard models were used to investigate the relationship between exposure and incident dementia with adjustment for covariates. Subgroup analyses were performed to assess the impact of severity of HL (unilateral vs. bilateral HADs; e.g., dose-response gradient) and the duration of exposure (2007–2010 vs. 2010–2013 vs. 2013–2016; e.g., exposure-response gradient). The adjusted models included all covariates used in the unadjusted analyses. Statistical significance was reached at p < 0.05. There was no missing data or loss to follow up as all data were captured by the administrative databases. All analyses were performed at ICES using the SAS Enterprise Guide 7.15.

Cohort Characteristics

Baseline subject characteristics are shown in Table 1. Using ADP claims, 1,262,295 participants were identified, with 257,285 patients in the cohort with ADP claims for HADs (ADP claimants) and 1,005,010 matched controls. The median ages were 72 years (Q1–Q3, 63–80) in the ADP cohort and 69 years (Q1–Q3, 60–77) in the control group. In the ADP cohort, 131,636 (51.16%) were female; 57,429 (22.32%) were in the highest income quintile, and 63,666 (24.75%) had diabetes. Diabetes was the most common health condition, followed by chronic obstructive pulmonary disease (20.45%), asthma (14.36%), and acute myocardial infarction (10.23%).

Table 1.

Demographics and general health characteristics of matched patients with hearing aid devices

Patient characteristicsAny hearing deviceWeighted standardized difference
total(N = 1,262,295)yes(N = 257,285)no(N = 1,005,010)
frequencypercentfrequencypercentfrequencypercent
Age 
 Mean±SD 68.79±11.84 70.91±12.00 68.24±11.74 0.000 
 Median (IQR) 69 (61–78) 72 (63–80) 69 (60–77)  
Age (categorized) 
 40–49 82,717 6.55 13,841 5.38 68,876 6.85 0.000 
 50–59 201,361 15.95 33,747 13.12 167,614 16.68 0.000 
 60–69 356,700 28.26 62,258 24.20 294,442 29.30 0.000 
 70+ 621,517 49.24 147,439 57.31 474,078 47.17 0.000 
Sex 
 Female 627,975 49.75 131,636 51.16 496,339 49.39 0.000 
 Male 634,320 50.25 125,649 48.84 508,671 50.61  
Neighborhood income quintile 
 1 (Lowest) 235,443 18.65 47,306 18.39 188,137 18.72 0.021 
 2 255,490 20.24 50,106 19.47 205,384 20.44 0.032 
 3 249,134 19.74 49,838 19.37 199,296 19.83 0.011 
 4 255,298 20.22 52,606 20.45 202,692 20.17 0.015 
 5 (Highest) 266,930 21.15 57,429 22.32 209,501 20.85 0.047 
Place of residence (LHIN) 
 01 68,562 5.43 15,452 6.01 53,110 5.28 0.030 
 02 102,284 8.10 24,537 9.54 77,747 7.74 0.063 
 03 66,435 5.26 15,664 6.09 50,771 5.05 0.046 
 04 154,057 12.20 33,060 12.85 120,997 12.04 0.020 
 05 61,522 4.87 10,156 3.95 51,366 5.11 0.052 
 06 92,442 7.32 17,238 6.70 75,204 7.48 0.026 
 07 99,903 7.91 17,153 6.67 82,750 8.23 0.063 
 08 150,379 11.91 25,599 9.95 124,780 12.42 0.080 
 09 150,445 11.92 30,162 11.72 120,283 11.97 0.008 
 10 58,814 4.66 13,499 5.25 45,315 4.51 0.034 
 11 118,044 9.35 26,835 10.43 91,209 9.08 0.049 
 12 47,937 3.80 10,418 4.05 37,519 3.73 0.019 
 13 66,636 5.28 12,238 4.76 54,398 5.41 0.029 
 14 24,835 1.97 5,274 2.05 19,561 1.95 0.008 
Chronic comorbidities 
 Asthma 156,266 12.38 36,941 14.36 119,325 11.87 0.069 
 COPD 224,889 17.82 52,603 20.45 172,286 17.14 0.048 
 CHF 90,587 7.18 21,679 8.43 68,908 6.86 0.003 
 Stroke 39,131 3.10 8,779 3.41 30,352 3.02 0.008 
 AMI 113,051 8.96 26,321 10.23 86,730 8.63 0.020 
 Diabetes 291,066 23.06 63,666 24.75 227,400 22.63 0.019 
Patient characteristicsAny hearing deviceWeighted standardized difference
total(N = 1,262,295)yes(N = 257,285)no(N = 1,005,010)
frequencypercentfrequencypercentfrequencypercent
Age 
 Mean±SD 68.79±11.84 70.91±12.00 68.24±11.74 0.000 
 Median (IQR) 69 (61–78) 72 (63–80) 69 (60–77)  
Age (categorized) 
 40–49 82,717 6.55 13,841 5.38 68,876 6.85 0.000 
 50–59 201,361 15.95 33,747 13.12 167,614 16.68 0.000 
 60–69 356,700 28.26 62,258 24.20 294,442 29.30 0.000 
 70+ 621,517 49.24 147,439 57.31 474,078 47.17 0.000 
Sex 
 Female 627,975 49.75 131,636 51.16 496,339 49.39 0.000 
 Male 634,320 50.25 125,649 48.84 508,671 50.61  
Neighborhood income quintile 
 1 (Lowest) 235,443 18.65 47,306 18.39 188,137 18.72 0.021 
 2 255,490 20.24 50,106 19.47 205,384 20.44 0.032 
 3 249,134 19.74 49,838 19.37 199,296 19.83 0.011 
 4 255,298 20.22 52,606 20.45 202,692 20.17 0.015 
 5 (Highest) 266,930 21.15 57,429 22.32 209,501 20.85 0.047 
Place of residence (LHIN) 
 01 68,562 5.43 15,452 6.01 53,110 5.28 0.030 
 02 102,284 8.10 24,537 9.54 77,747 7.74 0.063 
 03 66,435 5.26 15,664 6.09 50,771 5.05 0.046 
 04 154,057 12.20 33,060 12.85 120,997 12.04 0.020 
 05 61,522 4.87 10,156 3.95 51,366 5.11 0.052 
 06 92,442 7.32 17,238 6.70 75,204 7.48 0.026 
 07 99,903 7.91 17,153 6.67 82,750 8.23 0.063 
 08 150,379 11.91 25,599 9.95 124,780 12.42 0.080 
 09 150,445 11.92 30,162 11.72 120,283 11.97 0.008 
 10 58,814 4.66 13,499 5.25 45,315 4.51 0.034 
 11 118,044 9.35 26,835 10.43 91,209 9.08 0.049 
 12 47,937 3.80 10,418 4.05 37,519 3.73 0.019 
 13 66,636 5.28 12,238 4.76 54,398 5.41 0.029 
 14 24,835 1.97 5,274 2.05 19,561 1.95 0.008 
Chronic comorbidities 
 Asthma 156,266 12.38 36,941 14.36 119,325 11.87 0.069 
 COPD 224,889 17.82 52,603 20.45 172,286 17.14 0.048 
 CHF 90,587 7.18 21,679 8.43 68,908 6.86 0.003 
 Stroke 39,131 3.10 8,779 3.41 30,352 3.02 0.008 
 AMI 113,051 8.96 26,321 10.23 86,730 8.63 0.020 
 Diabetes 291,066 23.06 63,666 24.75 227,400 22.63 0.019 

SD, standard deviation; IQR, interquartile range; LHIN, Local Health Integrated Network; COPD, chronic obstructive pulmonary disease; CHF, congestive heart failure; AMI, acute myocardial infarction.

Incidence of Dementia

87,280 (6.91%) were diagnosed with dementia, with 22,549 (8.76%) in the ADP cohort and 64,731 (6.44%) in the control group. Of all the patients with ADP claims, 192,976 claimed bilateral HADs, and 64,309 claimed unilateral HADs. 17,059 (8.84%) patients with bilateral HADs and 5,490 (8.54%) with unilateral HADs are diagnosed with dementia as shown in Table 2. Dementia incidence rates (per 1,000 person-years) were 19.51 (95% confidence interval [CI]: 19.26–19.77) and 14.15 (95% CI: 14.04–14.26) for the ADP claimants and matched controls, respectively. The probability of dementia-free survival was lower in ADP claimants than in controls (5-year survival: 90.68% [95% CI: 90.56–90.79%] vs. 93.17% [95% CI: 93.12–93.22%]) (shown in Fig. 1). The risk of dementia was 10% higher among patients with ADP claims as compared to controls in adjusted analysis (hazard ratio [HR]: 1.10 [95% CI: 1.09–1.12, p < 0.001]). In subgroup analyses, the risk of dementia was 12% higher among those with bilateral HADs (HR: 1.12 [95% CI: 1.10–1.14, p < 0.001]) and 7% higher among those with unilateral HADs compared to the control group (HR: 1.07 [95% CI: 1.04–1.11, p < 0.001]) (shown in Table 3).

Table 2.

Dementia incident outcomes of matched patients with hearing aid devices

Patient characteristicsTotal(N = 1,262,295)Yes(N = 257,285)No(N = 1,005,010)Weighted standardized difference
frequencypercentfrequencypercentfrequencypercent
Any hearing device 
 Dementia incident 87,280 6.91 22,549 8.76 64,731 6.44 0.015 
Patient characteristicsTotal(N = 1,262,295)Yes(N = 257,285)No(N = 1,005,010)Weighted standardized difference
frequencypercentfrequencypercentfrequencypercent
Any hearing device 
 Dementia incident 87,280 6.91 22,549 8.76 64,731 6.44 0.015 
Total(N = 943,047)Yes(N = 192,976)No(N = 750,071)
frequencypercentfrequencypercentfrequencypercent
Bilateral hearing device 
 Dementia incident 65,632 6.96 17,059 8.84 48,573 6.48 0.017 
Total(N = 943,047)Yes(N = 192,976)No(N = 750,071)
frequencypercentfrequencypercentfrequencypercent
Bilateral hearing device 
 Dementia incident 65,632 6.96 17,059 8.84 48,573 6.48 0.017 
Total(N = 319,248)Yes(N = 64,309)No(N = 254,939)
frequencypercentfrequencypercentfrequencypercent
Unilateral hearing device 
 Dementia incident 21,648 6.78 5,490 8.54 16,158 6.34 0.009 
Total(N = 319,248)Yes(N = 64,309)No(N = 254,939)
frequencypercentfrequencypercentfrequencypercent
Unilateral hearing device 
 Dementia incident 21,648 6.78 5,490 8.54 16,158 6.34 0.009 
Fig. 1.

Kaplan-Meier plot for dementia incident outcomes of matched patients with hearing aid devices.

Fig. 1.

Kaplan-Meier plot for dementia incident outcomes of matched patients with hearing aid devices.

Close modal
Table 3.

Unadjusted and adjusted Cox regression for the development of dementia

Patient characteristicsOutcome: Dementia incidenta
unadjusted analysisadjusted analysisb
HR (95% CI)p valueHR (95% CI)p value
Any hearing device 
 Yes versus no 1.38 (1.36–1.40) <0.001 1.10 (1.09–1.12) <0.001 
Bilateral hearing device 
 Yes versus no 1.38 (1.36–1.41) <0.001 1.12 (1.10–1.14) <0.001 
Unilateral hearing device 
 Yes versus no 1.37 (1.33–1.41) <0.001 1.07 (1.04–1.11) <0.001 
Study period: April 1, 2007-March 31, 2010 
 Yes versus no 1.26 (1.23–1.29) <0.001 1.03 (1.01–1.06) 0.014 
Study period: April 1, 2010-March 31, 2013 
 Yes versus no 1.41 (1.38–1.45) <0.001 1.12 (1.09–1.15) <0.001 
Study period: April 1, 2013-March 31, 2016 
 Yes versus no 1.50 (1.46–1.54) <0.001 1.19 (1.16–1.23) <0.001 
Age (categorized) 
 40–49 Reference  Reference  
 50–59 2.42 (2.04–2.87) <0.001 2.36 (1.99–2.80) <0.001 
 60–69 10.22 (8.71–11.99) <0.001 9.56 (8.15–11.21) <0.001 
 70+ 82.63 (70.56–96.76) <0.001 64.83 (55.35–75.92) <0.001 
Sex 
 Female Reference  Reference  
 Male 0.69 (0.68–0.70) <0.001 0.87 (0.86–0.88) <0.001 
Neighborhood income quintile 
 1 (Lowest) Reference  Reference  
 2 0.90 (0.88–0.92) <0.001 0.95 (0.93–0.97) <0.001 
 3 0.80 (0.78–0.82) <0.001 0.90 (0.88–0.92) <0.001 
 4 0.74 (0.73–0.76) <0.001 0.87 (0.85–0.89) <0.001 
 5 (Highest) 0.71 (0.69–0.72) <0.001 0.85 (0.83–0.86) <0.001 
Chronic conditions 
 Asthma (yes vs. no) 1.15 (1.12–1.17) <0.001 0.95 (0.93–0.97) <0.001 
 COPD (yes vs. no) 1.77 (1.74–1.79) <0.001 1.18 (1.16–1.20) <0.001 
 CHF (yes vs. no) 2.85 (2.80–2.91) <0.001 1.35 (1.32–1.38) <0.001 
 Stroke (yes vs. no) 3.22 (3.14–3.30) <0.001 1.71 (1.67–1.76) <0.001 
 AMI (yes vs. no) 1.85 (1.82–1.89) <0.001 1.09 (1.07–1.11) <0.001 
 Diabetes (yes vs. no) 1.58 (1.56–1.61) <0.001 1.12 (1.11–1.14) <0.001 
Patient characteristicsOutcome: Dementia incidenta
unadjusted analysisadjusted analysisb
HR (95% CI)p valueHR (95% CI)p value
Any hearing device 
 Yes versus no 1.38 (1.36–1.40) <0.001 1.10 (1.09–1.12) <0.001 
Bilateral hearing device 
 Yes versus no 1.38 (1.36–1.41) <0.001 1.12 (1.10–1.14) <0.001 
Unilateral hearing device 
 Yes versus no 1.37 (1.33–1.41) <0.001 1.07 (1.04–1.11) <0.001 
Study period: April 1, 2007-March 31, 2010 
 Yes versus no 1.26 (1.23–1.29) <0.001 1.03 (1.01–1.06) 0.014 
Study period: April 1, 2010-March 31, 2013 
 Yes versus no 1.41 (1.38–1.45) <0.001 1.12 (1.09–1.15) <0.001 
Study period: April 1, 2013-March 31, 2016 
 Yes versus no 1.50 (1.46–1.54) <0.001 1.19 (1.16–1.23) <0.001 
Age (categorized) 
 40–49 Reference  Reference  
 50–59 2.42 (2.04–2.87) <0.001 2.36 (1.99–2.80) <0.001 
 60–69 10.22 (8.71–11.99) <0.001 9.56 (8.15–11.21) <0.001 
 70+ 82.63 (70.56–96.76) <0.001 64.83 (55.35–75.92) <0.001 
Sex 
 Female Reference  Reference  
 Male 0.69 (0.68–0.70) <0.001 0.87 (0.86–0.88) <0.001 
Neighborhood income quintile 
 1 (Lowest) Reference  Reference  
 2 0.90 (0.88–0.92) <0.001 0.95 (0.93–0.97) <0.001 
 3 0.80 (0.78–0.82) <0.001 0.90 (0.88–0.92) <0.001 
 4 0.74 (0.73–0.76) <0.001 0.87 (0.85–0.89) <0.001 
 5 (Highest) 0.71 (0.69–0.72) <0.001 0.85 (0.83–0.86) <0.001 
Chronic conditions 
 Asthma (yes vs. no) 1.15 (1.12–1.17) <0.001 0.95 (0.93–0.97) <0.001 
 COPD (yes vs. no) 1.77 (1.74–1.79) <0.001 1.18 (1.16–1.20) <0.001 
 CHF (yes vs. no) 2.85 (2.80–2.91) <0.001 1.35 (1.32–1.38) <0.001 
 Stroke (yes vs. no) 3.22 (3.14–3.30) <0.001 1.71 (1.67–1.76) <0.001 
 AMI (yes vs. no) 1.85 (1.82–1.89) <0.001 1.09 (1.07–1.11) <0.001 
 Diabetes (yes vs. no) 1.58 (1.56–1.61) <0.001 1.12 (1.11–1.14) <0.001 

HR, hazard ratio; CI, confidence interval; LHIN, Local Health Integrated Network; COPD, chronic obstructive pulmonary disease; CHF, congestive heart failure; AMI, acute myocardial infarction.

aDementia incident is measured within 5-year post-index date.

bAdjusted model contains all variables used in the unadjusted analyses.

Study Period and Risk of Dementia

83,514, 86,864, and 86,907 patients were identified through ADP claims from April 1, 2007, to March 31, 2010, April 1, 2010, to March 31, 2013, and April 1, 2013, to March 31, 2016, respectively. Of these patients, 8,226 (9.85%) in 2007–2010, 7,559 (8.70%) in 2010–2013, and 6,764 (7.78%) in 2013–2016 were diagnosed with dementia. Of the study periods analyzed, the risk of dementia was 3% higher during 2007–2010 (HR: 1.03 [95% CI: 1.01–1.06, p = 0.014]), 12% higher during 2010–2013 (HR: 1.12 [95% CI: 1.09–1.15, p < 0.001]) and 19% higher during 2013–2016 (HR: 1.19 [95% CI: 1.16–1.23, p < 0.001]).

This large population-based cohort study examines the risk of incident dementia in ADP claimants compared with matched controls over 14 years in a comprehensive single-payer system. The findings indicate a higher incidence of dementia diagnosis among those with HL as identified via ADP claims compared to matched controls. We demonstrate a higher hazard of incident dementia in older adults with HAD claims with a dose-dependent relationship. Moreover, our results indicate an exposure-response gradient with an increasing magnitude of dementia risk over time. Despite HL being a potentially modifiable risk factor for dementia, the rate of HAD use remains suboptimal [21].

A greater risk of dementia diagnosis was observed in this study in patients with ADP claims (19.51 per 1,000 person-years) when compared to matched controls (14.15 per person-years). Our findings agree with similar studies that have demonstrated HL as a risk factor of incident dementia [11, 22, 23]. Due to the lack of audiometric data to quantify the severity of HL, bilateral versus unilateral HAD claims were used as a marker for HL severity. We demonstrate a dose-dependent relationship whereby individuals with bilateral HAD claims had a higher hazard of incident dementia. Bilateral HAD claims were associated with 1.12 times the hazard of incident dementia, adjusting for covariates. Unilateral HAD claims were associated with 1.07 times the hazard of incident dementia, adjusting for covariates. Overall HAD claims were associated with 1.1 times the hazard of incident dementia, adjusting for covariates. Our findings agree with the literature indicating a dose-dependent relationship with the degree of cognitive deficit being significantly associated with the severity of HL [24, 25].

A recent national study found an 11.46% prevalence of dementia in hearing aid device (HAD) users, with a dose-response relationship between HL severity and dementia prevalence [26]. Our results are consistent with these findings. There are multiple hypotheses regarding the potential link between HL and dementia [13, 15]. The cognitive load hypothesis suggests that individuals with HL may allocate cognitive resources to auditory perceptual processing, leading to increased cognitive burden during listening [13]. Additionally, HL may be linked to social isolation, causing individuals to withdraw from social activities [15]. Taken together, the use of HADs may mitigate cognitive burden from listening, provide sensory stimulation to prevent prolonged deprivation, and enhance social connectedness.

The demonstrated risk of dementia incidence in patients with access to hearing aids is somewhat counterintuitive and contrary to other studies. When analyzing the temporality of ADP claims, we observe an increasing magnitude of dementia risk over time. HAD claims between 2007–2010 were associated with 1.03 times the hazard of incident dementia, 2010–2013 were associated with 1.12 times the hazard of incident dementia, and 2013–2016 were associated with 1.19 times the hazard of incident dementia. This analysis suggests an exposure-response gradient, with a longer duration of exposure to HADs associated with decreased risk of incident dementia. HL remains one of the most untreated conditions making it a compelling target in preventative strategies for dementia [27]. While randomized controlled trials are needed to demonstrate the efficacy of hearing aids in preventing the development of dementia in those with HL, a recent meta-analysis demonstrated that use of HADs was found to be significantly associated with a 19% decrease in the hazards of long-term cognitive decline [28]. In addition, studies demonstrate delayed diagnosis of dementia for up to 3–4 years in HAD users [28, 29]. Furthermore, our study followed participants for a period of 5 years and indicate an accelerated rate of cognitive decline in individuals with HL. This finding is consistent with another study which demonstrated a 30–40% accelerated rate of cognitive decline and a 24% increased risk for incident cognitive impairment over a 6-year period in individuals with HL [27]. However, our study revealed a lower magnitude of increased risk, potentially suggesting that the use of HADs may have protective effects in slowing the progression of dementia onset. Although beneficial, it is important to acknowledge that the use of HADs is influenced by other factors such as nonadherence, cognitive and functional restrictions, and perceived lack of benefits [30, 31]. These factors can influence long-term benefits and should be explored with the patients and caregivers.

Our study has several strengths. The comprehensive single-payer databases available in Ontario allowed for an objective population-based measure with large sample sizes. Our study overcomes several limitations of previous studies since it has large cohorts comprising almost the entire adult population in Ontario, the most populous province in Canada, and lagged length of intervention to reduce concerns about reverse causality. Given the single-payer nature of the system, the data collected is unbiased and is minimally affected by discrepancies in socioeconomic status. With demographic characteristics like the USA and many European countries, findings from this study are generalizable to populations in many other regions despite the varying hearing care models.

Our study’s limitations are intrinsic to observational studies that use administrative data. Our data are limited to patients who claimed HADs and did not capture everyone with a diagnosis of HL. Furthermore, HADs are cost-prohibitive even with ADP assistance and may not be accessible to everyone. The database lacks audiometric information that would allow for the identification of true duration and severity of HL. Using bilateral versus unilateral HADs as a surrogate for HL severity can limit interpretation. In addition, the insidious nature of dementia onset makes the identification of diagnosis difficult. Another limitation of the study lies in the lack of information on confounding risk factors including education level and lifestyle behaviors which make it difficult to exclude residual statistical confounding in the interpretation of the results. Moreover, it is crucial to recognize that the utilization of HADs may be affected by various other factors, including nonadherence, functional limitations, and perceived lack of benefits. These factors, which are not reflected in administrative data, can potentially influence the interpretation of study results. Given the observational study design, associations are identified but do not necessarily reflect causation. Future studies should employ standardized diagnostic codes to identify more patients and allow for robust characterization of HL. Randomized controlled trials are required to assess whether treating HL can reduce incident dementia [32].

HL is an invisible disability that has numerous consequences for health and quality of life. Additionally, the development of dementia has psychological, social, and economic impacts on patients, caregivers, and society. Although dementia mainly affects older individuals, it is not an inevitable consequence of aging. By targeting modifiable risk factors like HL, the risk of dementia can be lowered [21]. Our findings provide insights into the incidence of dementia in persons with HL. Given the potential role of HL in dementia risk, understanding the effect of early hearing interventions merits further investigation.

  1. 1.

    This study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC). This document used data adapted from the Statistics Canada Postal CodeOM Conversion File, which is based on data licensed from Canada Post Corporation, and/or data adapted from the Ontario Ministry of Health Postal Code Conversion File, which contains data copied under license from ©Canada Post Corporation and Statistics Canada. Part of this material is based on data and information compiled and provided by: MOH, CIHI. The analyses, conclusions, opinions, and statements expressed herein are solely those of the authors and do not reflect those of the funding or data resources; no endorsement is intended or should be inferred.

  2. 2.

    Parts of this material are based on data and/or information compiled and provided by CIHI. However, the analyses, conclusions, opinions, and statements expressed in the material are those of the author(s), and not necessarily those of CIHI.

  3. 3.

    The dataset was cut at ICES Queen’s (Kingston, ON). The dataset creation and data analysis were performed at ICES Queen’s by Dr. Paul Nguyen.

This study was performed in accordance with Institutional Ethical Guidelines and the Declaration of Helsinki. A waiver of informed consent was granted based on the Ontario’s Personal Health Information Protection Act (PHIPA), Section 45, with ethics approval by Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board (HSREB), project #6025803. Section 45 of PHIPA authorizes ICES to collect personal health information, without consent, for the purpose of analysis or compiling statistical information with respect to the management of, evaluation or monitoring of, the allocation of resources to or planning for all or part of the health system.

No conflicting relationships exist for any author. No financial interests exist for any author.

Funding was provided by the SEAMO New Clinician Scientist Program Research Establishment Funding. Role of the funders is financial support only.

Keshinisuthan Kirubalingam: conducted analyses, interpreted results, drafted, and critically revised the manuscript, and approved the final and submitted version. Paul Nguyen: acquired data, conducted analyses, critically revised the manuscript, and approved the final and submitted version. Daniel Newsted, Allison De La Lis, and Sudeep Gill: interpreted results, critically revised the manuscript, and approved the final and submitted version. Jason A. Beyea: conceived and designed the study, interpreted results, critically revised the manuscript, and approved the final and submitted version.

The dataset from this study is held securely in coded form at ICES. While legal data sharing agreements between ICES and data providers (e.g., healthcare organizations and government) prohibit ICES from making the dataset publicly available, access may be granted to those who meet prespecified criteria for confidential access, available at www.ices.on.ca/DAS (email: das@ices.on.ca). The full dataset creation plan and underlying analytic code are available from the authors upon request, understanding that the computer programs may rely upon coding templates or macros that are unique to ICES and are therefore either inaccessible or may require modification.

1.
Feder
KP
,
Michaud
D
,
McNamee
J
,
Fitzpatrick
E
,
Ramage-Morin
P
,
Beauregard
Y
.
Prevalence of hearing loss among a representative sample of Canadian children and adolescents, 3 to 19 years of age
.
Ear Hear
.
2017
;
38
(
1
):
7
20
.
2.
Lin
FR
.
Hearing loss and cognition among older adults in the United States
.
J Gerontol A Biol Sci Med Sci
.
2011
;
66
(
10
):
1131
6
.
3.
Lin
FR
,
Ferrucci
L
,
Metter
EJ
,
An
Y
,
Zonderman
AB
,
Resnick
SM
.
Hearing loss and cognition in the baltimore longitudinal study of aging
.
Neuropsychology
.
2011
;
25
(
6
):
763
70
.
4.
American Psychiatric Association
.
Diagnostic and statistical manual of mental disorders
5th ed.
Text rev
Washington (DC)
American Psychiatric Association
2022
5.
The Government of Canada
Mapping connections an understanding of neurological conditions
.
2014
. [Internet]. Phac-aspc.gc.ca. [cited 2023 Jan 11] http://www.phac-aspc.gc.ca/publicat/cd-mc/mc-ec/index-eng.php.
6.
O’Gorman
C
,
Lucas
R
,
Taylor
B
.
Environmental risk factors for multiple sclerosis: a review with a focus on molecular mechanisms
.
Int J Mol Sci
.
2012
;
13
(
9
):
11718
52
.
7.
Wirdefeldt
K
,
Adami
H-O
,
Cole
P
,
Trichopoulos
D
,
Mandel
J
.
Epidemiology and etiology of Parkinson’s disease: a review of the evidence
.
Eur J Epidemiol
.
2011
26
Suppl 1
S1
58
.
8.
Baumgart
M
,
Snyder
HM
,
Carrillo
MC
,
Fazio
S
,
Kim
H
,
Johns
H
.
Summary of the evidence on modifiable risk factors for cognitive decline and dementia: a population-based perspective
.
Alzheimers Dement
.
2015
;
11
(
6
):
718
26
.
9.
Loughrey
DG
,
Kelly
ME
,
Kelley
GA
,
Brennan
S
,
Lawlor
BA
.
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
(
2
):
115
26
.
10.
Byun
H
,
Chung
JH
,
Lee
SH
,
Kim
EM
,
Kim
I
.
Dementia in a hearing-impaired population according to hearing aid use: a nationwide population-based study in Korea
.
Ear Hear
.
2022
;
43
(
6
):
1661
8
.
11.
Davies
HR
,
Cadar
D
,
Herbert
A
,
Orrell
M
,
Steptoe
A
.
Hearing impairment and incident dementia: findings from the English longitudinal study of ageing
.
J Am Geriatr Soc
.
2017
;
65
(
9
):
2074
81
.
12.
Livingston
G
,
Huntley
J
,
Sommerlad
A
,
Ames
D
,
Ballard
C
,
Banerjee
S
.
Dementia prevention, intervention, and care: 2020 report of the Lancet Commission
.
Lancet
.
2020
;
396
(
10248
):
413
46
.
13.
Griffiths
TD
,
Lad
M
,
Kumar
S
,
Holmes
E
,
McMurray
B
,
Maguire
EA
.
How can hearing loss cause dementia
.
Neuron
.
2020
;
108
(
3
):
401
12
.
14.
Manohar
SG
,
Zokaei
N
,
Fallon
SJ
,
Vogels
TP
,
Husain
M
.
Neural mechanisms of attending to items in working memory
.
Neurosci Biobehav Rev
.
2019
;
101
:
1
12
.
15.
Amieva
H
,
Ouvrard
C
,
Meillon
C
,
Rullier
L
,
Dartigues
J-F
.
Death, depression, disability, and dementia associated with self-reported hearing problems: a 25-year study
.
J Gerontol A Biol Sci Med Sci
.
2018
;
73
(
10
):
1383
9
.
16.
Yeo
BSY
,
Song
HJJMD
,
Toh
EMS
,
Ng
LS
,
Ho
CSH
,
Ho
R
.
Association of hearing aids and cochlear implants with cognitive decline and dementia: a Systematic Review and meta-analysis
.
JAMA Neurol
.
2023 Feb 1
80
2
134
41
.
17.
Castiglione
A
,
Benatti
A
,
Velardita
C
,
Favaro
D
,
Padoan
E
,
Severi
D
.
Aging, cognitive decline and hearing loss: effects of auditory rehabilitation and training with hearing aids and cochlear implants on cognitive function and depression among older adults
.
Audiol Neurootol
.
2016
21
Suppl. 1
21
8
.
18.
Government of Canada, Canada S
Annual demographic estimates: Canada
. provinces and territories [Internet]. Statcan.gc.ca. 2022 [cited 2023 Jan 11] https://www150.statcan.gc.ca/n1/en/catalogue/91-215-X.
19.
Government of Ontario
Hearing Devices [Internet]
Ontario, CA
. [cited 2023 Jan 11]. doi: https://www.ontario.ca/page/hearing-devices.
20.
Jaakkimainen
RL
,
Bronskill
SE
,
Tierney
MC
,
Herrmann
N
,
Green
D
,
Young
J
.
Identification of physician-diagnosed alzheimer’s disease and related dementias in population-based administrative data: a validation study using family physicians’ electronic medical records
.
J Alzheimers Dis
.
2016
;
54
(
1
):
337
49
.
21.
Ford
AH
,
Hankey
GJ
,
Yeap
BB
,
Golledge
J
,
Flicker
L
,
Almeida
OP
.
Hearing loss and the risk of dementia in later life
.
Maturitas
.
2018
;
112
:
1
11
.
22.
Gurgel
RK
,
Ward
PD
,
Schwartz
S
,
Norton
MC
,
Foster
NL
,
Tschanz
JT
.
Relationship of hearing loss and dementia: a prospective, population-based study
.
Otol Neurotol
.
2014
;
35
(
5
):
775
81
.
23.
Chern
A
,
Sharma
RK
,
Golub
JS
.
Hearing loss and incident dementia: claims data from the New York SPARCS database
.
Otol Neurotol
.
2022
;
43
(
1
):
36
41
.
24.
Taljaard
DS
,
Olaithe
M
,
Brennan-Jones
CG
,
Eikelboom
RH
,
Bucks
RS
.
The relationship between hearing impairment and cognitive function: a meta-analysis in adults
.
Clin Otolaryngol
.
2016
;
41
(
6
):
718
29
.
25.
Chern
A
,
Golub
JS
.
Age-related hearing loss and dementia
.
Alzheimer Dis Assoc Disord
.
2019
;
33
(
3
):
285
90
.
26.
Huang
AR
,
Jiang
K
,
Lin
FR
,
Deal
JA
,
Reed
NS
.
Hearing loss and dementia prevalence in older adults in the US
.
JAMA
.
2023
;
329
(
2
):
171
3
.
27.
Lin
FR
,
Yaffe
K
,
Xia
J
,
Xue
Q-L
,
Harris
TB
,
Purchase-Helzner
E
.
Hearing loss and cognitive decline in older adults
.
JAMA Intern Med
.
2013
;
173
(
4
):
293
9
.
28.
Mahmoudi
E
,
Basu
T
,
Langa
K
,
McKee
MM
,
Zazove
P
,
Alexander
N
.
Can hearing aids delay time to diagnosis of dementia, depression, or falls in older adults
.
J Am Geriatr Soc
.
2019
;
67
(
11
):
2362
9
.
29.
Bucholc
M
,
McClean
PL
,
Bauermeister
S
,
Todd
S
,
Ding
X
,
Ye
Q
.
Association of the use of hearing aids with the conversion from mild cognitive impairment to dementia and progression of dementia: a longitudinal retrospective study
.
Alzheimers Dement
.
2021
;
7
(
1
):
e12122
.
30.
McCormack
A
,
Fortnum
H
.
Why do people fitted with hearing aids not wear them
.
Int J Audiol
.
2013
;
52
(
5
):
360
8
.
31.
Jenstad
L
,
Moon
J
.
Systematic review of barriers and facilitators to hearing aid uptake in older adults
.
Audiol Res
.
2011
;
1
(
1
):
e25
.
32.
Deal
JA
,
Albert
MS
,
Arnold
M
,
Bangdiwala
SI
,
Chisolm
T
,
Davis
S
.
A randomized feasibility pilot trial of hearing treatment for reducing cognitive decline: results from the Aging and Cognitive Health Evaluation in Elders Pilot Study
.
Alzheimers Dement
.
2017
;
3
(
3
):
410
5
.