Redefining Hearing Loss in Older Adults: Implications for Psychiatry

OBJECTIVE: To assess whether the relationship between hearing and depressive symptoms is present among older adults classified as normal hearing (≤25 dB). DESIGN: Cross-sectional epidemiologic study embedded within a prospective cohort study (Hispanic Community Health Study) SETTING: Multi-centered at 4 US communities (New York, Chicago, Miami, San Diego) PARTICIPANTS: Adults ≥50 years old (n=5,499) with normal hearing or hearing loss. MEASUREMENTS: The primary exposure was hearing, defined continuously by the 4-frequency pure-tone average on audiometry (mean hearing threshold, in dB, at 500, 1000, 2000, and 4000 Hz pitch). Hearing was additionally categorized into normal hearing (≤25 dB) and hearing loss (>25 dB). The main outcome was depressive symptoms, measured with the Center for Epidemiologic Studies Depression Scale-10 (CESD-10). Depressive symptoms were defined both continuously and binarily (where CESD-10 ≥ 10 was categorized as clinically significant depressive symptoms). Multivariable linear, logistic, and generalized additive modeling (GAM) regressions were performed. RESULTS: Worse hearing was related to higher depressive symptoms among those with normal hearing in GAM regression. Among those with normal hearing, the CESD-10 score increased by 1.04 points (95% CI = 0.70, 1.37) for every 10 dB decrease in hearing, adjusting for age, gender, education, cardiovascular disease, and hearing aid use. Among those with hearing loss, the CESD-10 score increased by 0.62 points (95% CI = 0.23, 1.01) for every 10 dB decrease in hearing, adjusting for the same confounders. Similar findings were noted when the outcome was clinically significant depressive symptoms (adjusted OR=1.28 [1.14, 1.44] in normal hearing versus 1.26 [1.11, 1.44] in HL). In certain sensitivity analyses, the relationship between hearing and depressive symptoms was significantly stronger among those with normal hearing than in those with hearing loss. CONCLUSION: The relationship between hearing and clinically significant depressive symptoms is present among older adults with normal hearing (