Incident Hearing Loss and Comorbidity: A Longitudinal Administrative Claims Study

• Published in: JAMA otolaryngology-- head & neck surgery Vol. 145; no. 1; p. 36
• Main Authors: Deal, Jennifer A, Reed, Nicholas S, Kravetz, Alexander D, Weinreich, Heather, Yeh, Charlotte, Lin, Frank R, Altan, Aylin
• Format: Journal Article
• Language: English
• Published: United States 01.01.2019
• Subjects: Aged; Aged, 80 and over; Case-Control Studies; Comorbidity; Databases, Factual; Female; Follow-Up Studies; Hearing Loss - epidemiology; Humans; Incidence; Male; Middle Aged; Propensity Score; Retrospective Studies; United States - epidemiology; United States
• ISSN: 2168-619X
• DOI: 10.1001/jamaoto.2018.2876

Because hearing loss is highly prevalent and treatable, determining its association with morbidity has major public health implications for disease prevention and the maintenance of health in adults with hearing loss. To investigate the association between the diagnosis of incident hearing loss and medical comorbidities in adults 50 years or older. Retrospective, propensity-matched cohort study using administrative claims data from commercially insured and Medicare Advantage members in a geographically diverse US health plan. Adults 50 years or older with claims for services rendered from January 1, 2000, to December 31, 2016, were observed for 2 (n = 154 414), 5 (n = 44 852), and 10 (n = 4728) years. This research was conceptualized and data were analyzed between September 2016 and November 2017. A claim for incident hearing loss is defined as 2 claims for hearing loss within 2 consecutive years without evidence of hearing device use, excluding claims for sudden hearing loss or hearing loss secondary to medical conditions. Incident claims for dementia, depression, accidental falls, nonvertebral fractures, acute myocardial infarction, and stroke. After cohort matching, 48% of participants were women (n = 74 464), 61% were white (n = 93 442), and 31% (n = 48 056) were Medicare Advantage insured, with a mean (SD) age of 64 (10) years. In a multivariate-adjusted modified Poisson regression with robust standard errors, relative associations were strongest for dementia (relative risk at 5 years, 1.50; 95% CI, 1.38-1.64) and depression (relative risk at 5 years, 1.41; 95% CI, 1.26-1.58). The absolute risk of all outcomes was greater in persons with hearing loss than in those without hearing loss at all times, with the greatest risk difference observed at 10 years for all outcomes. The 10-year risk attributable to hearing loss was 3.20 per 100 persons (95% CI, 1.76-4.63) for dementia, 3.57 per 100 persons (95% CI, 1.67-5.47) for falls, and 6.88 per 100 persons (95% CI, 4.62-9.14) for depression. In this large observational study using administrative claims data, incident untreated hearing loss was associated with greater incident morbidity than no hearing loss across a range of health conditions. Future studies are needed to elucidate the mechanisms underlying these associations and to determine if treatment for hearing loss could reduce the risk of comorbidity.