Hearing Impairment and Allostatic Load in Older Adults

Abstract

Importance: Allostatic load, the cumulative strain that results from the chronic stress response, is associated with poor health outcomes. Increased cognitive load and impaired communication associated with hearing loss could potentially be associated with higher allostatic load, but few studies to date have quantified this association.

Objective: To investigate if audiometric hearing loss is associated with allostatic load and evaluate if the association varies by demographic factors.

Design, setting, participants: This cross-sectional survey used nationally representative data from the National Health and Nutrition Examination Survey. Audiometric testing was conducted from 2003 to 2004 (ages 20-69 years) and 2009 to 2010 (70 years or older). The study was restricted to participants aged 50 years or older, and the analysis was stratified based on cycle. The data were analyzed between October 2021 and October 2022.

Exposure: A 4-frequency (0.5-4.0 kHz) pure tone average was calculated in the better-hearing ear and modeled continuously and categorically (<25 dB hearing level [dB HL], no hearing loss; 26-40 dB HL, mild hearing loss; ≥41 dB HL, moderate or greater hearing loss).

Main outcome and measures: Allostatic load score (ALS) was defined using laboratory measurements of 8 biomarkers (systolic/diastolic blood pressure, body mass index [calculated as weight in kilograms divided by height in meters squared], and total serum and high-density lipoprotein cholesterol, glycohemoglobin, albumin, and C-reactive protein levels). Each biomarker was assigned a point if it was in the highest risk quartile based on statistical distribution and then summed to yield the ALS (range, 0-8). Linear regression models adjusted for demographic and clinical covariates. Sensitivity analysis included using clinical cut points for ALS and subgroup stratification.

Results: In 1412 participants (mean [SD] age, 59.7 [5.9] years; 293 women [51.9%]; 130 [23.0%] Hispanic, 89 [15.8%] non-Hispanic Black, and 318 [55.3%] non-Hispanic White individuals), a modest association was suggested between hearing loss and ALS (ages 50-69 years: β = 0.19 [95% CI, 0.02-0.36] per 10 dB HL; 70 years or older: β = 0.10 [95% CI, 0.02-0.18] per 10 dB HL) among non-hearing aid users. Results were not clearly reflected in the sensitivity analysis with clinical cut points for ALS or modeling hearing loss categorically. Sex-based stratifications identified a stronger association among male individuals (men 70 years or older: β = 0.22 [95% CI, 0.12-0.32] per 10 dB HL; women: β = 0.08 [95% CI, -0.04 to 0.20] per 10 dB HL).

Conclusion and relevance: The study findings did not clearly support an association between hearing loss and ALS. While hearing loss has been shown to be associated with increased risk for numerous health comorbidities, its association with the chronic stress response and allostasis may be less than that of other health conditions.

Figure 1.. Derivation of the Analytic Sample and Distribution of Allostatic Load Score (ALS) Modeled Using Statistical Distribution Using Quartiles by National Health and Nutrition Examination Survey (NHANES) Cycle

The sample was limited to adults 50 years or older from the NHANES 2003 to 2004 (age 50-69 years) and 2009 to 2010 (70 years or older) cycles with complete ALS and covariate data.

Figure 2.. Multivariable-Adjusted Association Between Hearing Loss and Allostatic Load Score (ALS) Modeled by Statistical Distribution Using Quartiles and by Clinical Cut Points

Corresponding β coefficients (95% CI) displaying either continuous change in ALS per 10 dB hearing loss (HL) or categorical change in ALS in those with mild or moderate or greater HL vs no HL. Higher ALS scores indicate a greater association of stress with dysregulation. The sample was limited to adults 50 years or older from the National Health and Nutrition Examination Survey (NHANES) 2003 to 2004 (age 50-69 years) and 2009 to 2010 (70 years or older) cycles. PTA indicates pure tone average. ^aStatistical significance of the β (95% CI). ^bStatistical significance of the HL*hearing aid use interaction term at α = .05.

Figure 3.. Subgroup Analysis by Sex of the Multivariable-Adjusted Association Between Hearing Loss (HL) and Allostatic Load Score (ALS) Modeled by Statistical Distribution Using Quartiles

Corresponding β coefficients (95% CI) displaying either continuous change in ALS per 10 dB HL or categorical change in ALS in those with mild or moderate or greater HL vs no HL. Higher ALS scores indicate a greater association with stress dysregulation. The sample was limited to adults 50 years or older from the National Health and Nutrition Examination Survey (NHANES) 2003 to 2004 (age 50-69 years) and 2009 to 2010 (70 years or older) cycles. PTA indicates pure tone average. ^aStatistical significance of the β (95% CI). ^bStatistical significance of the HL*hearing aid use interaction term at α = .05.