Trends in Health Care Costs and Utilization Associated With Untreated Hearing Loss Over 10 Years | Otolaryngology | JAMA Otolaryngology–Head & Neck Surgery | JAMA Network
[Skip to Content]
[Skip to Content Landing]
Figure 1.  Difference in Unadjusted Mean Patient-Paid, Plan-Paid, and Total Costs for Patients With Untreated Hearing Loss vs Patients With No Hearing Loss
Difference in Unadjusted Mean Patient-Paid, Plan-Paid, and Total Costs for Patients With Untreated Hearing Loss vs Patients With No Hearing Loss

Error bars represent the 95% CI. HPP indicates health plan–paid costs; OOP, out-of-pocket costs.

Figure 2.  Unadjusted Difference in 30-Day Readmissions for Participants With Untreated Hearing Loss vs Participants With No Hearing Loss
Unadjusted Difference in 30-Day Readmissions for Participants With Untreated Hearing Loss vs Participants With No Hearing Loss

Bar graph represents the unadjusted difference in baseline incidence of 30-day readmissions in the 2- (risk difference [RD], 0.45; 95% CI, 0.31-0.58), 5- (RD, 0.82; 95% CI, 0.49-1.15), and 10- (RD, 0.45; 95% CI, 0.49-1.15) year cohorts with untreated and no hearing loss.

Table 1.  Participant Baseline Characteristics
Participant Baseline Characteristics
Table 2.  Postpropensity Score Matching, Unadjusted Health-Service Costs and Utilization for Untreated Hearing Loss vs No Hearing Loss Groups
Postpropensity Score Matching, Unadjusted Health-Service Costs and Utilization for Untreated Hearing Loss vs No Hearing Loss Groups
Table 3.  Difference in Unadjusted Mean Costs for Untreated Hearing Loss vs No Hearing Loss Groups
Difference in Unadjusted Mean Costs for Untreated Hearing Loss vs No Hearing Loss Groups
1.
Lin  FR, Niparko  JK, Ferrucci  L.  Hearing loss prevalence in the United States.  Arch Intern Med. 2011;171(20):1851-1852. doi:10.1001/archinternmed.2011.506PubMedGoogle ScholarCrossref
2.
Goman  AM, Reed  NS, Lin  FR.  Addressing estimated hearing loss in adults in 2060.  JAMA Otolaryngol Head Neck Surg. 2017;143(7):733-734. doi:10.1001/jamaoto.2016.4642PubMedGoogle ScholarCrossref
3.
Lin  FR, Yaffe  K, Xia  J,  et al; Health ABC Study Group.  Hearing loss and cognitive decline in older adults.  JAMA Intern Med. 2013;173(4):293-299. doi:10.1001/jamainternmed.2013.1868PubMedGoogle ScholarCrossref
4.
Lin  FR, Metter  EJ, O’Brien  RJ, Resnick  SM, Zonderman  AB, Ferrucci  L.  Hearing loss and incident dementia.  Arch Neurol. 2011;68(2):214-220. doi:10.1001/archneurol.2010.362PubMedGoogle ScholarCrossref
5.
Lin  FR, Ferrucci  L.  Hearing loss and falls among older adults in the United States.  Arch Intern Med. 2012;172(4):369-371. doi:10.1001/archinternmed.2011.728PubMedGoogle ScholarCrossref
6.
Wallhagen  MI, Strawbridge  WJ, Shema  SJ, Kurata  J, Kaplan  GA.  Comparative impact of hearing and vision impairment on subsequent functioning.  J Am Geriatr Soc. 2001;49(8):1086-1092. doi:10.1046/j.1532-5415.2001.49213.xPubMedGoogle ScholarCrossref
7.
Dalton  DS, Cruickshanks  KJ, Klein  BE, Klein  R, Wiley  TL, Nondahl  DM.  The impact of hearing loss on quality of life in older adults.  Gerontologist. 2003;43(5):661-668. doi:10.1093/geront/43.5.661PubMedGoogle ScholarCrossref
8.
Foley  DM, Frick  KD, Lin  FR.  Association between hearing loss and healthcare expenditures in older adults.  J Am Geriatr Soc. 2014;62(6):1188-1189. doi:10.1111/jgs.12864PubMedGoogle ScholarCrossref
9.
Genther  DJ, Betz  J, Pratt  S,  et al; Health, Aging and Body Composition Study.  Association between hearing impairment and risk of hospitalization in older adults.  J Am Geriatr Soc. 2015;63(6):1146-1152. doi:10.1111/jgs.13456PubMedGoogle ScholarCrossref
10.
Chien  W, Lin  FR.  Prevalence of hearing aid use among older adults in the United States.  Arch Intern Med. 2012;172(3):292-293. doi:10.1001/archinternmed.2011.1408PubMedGoogle ScholarCrossref
11.
National Academies of Sciences, Engineering, and Medicine.  Hearing Health Care for Adults: Priorities for Improving Access and Affordability. Washington, DC: The National Academies Press; 2016.
12.
President’s Council of Advisors on Science and Technology (US).  Aging America and Hearing Loss: Imperative of Improved Hearing Technologies. Washington, DC: Executive Office of the President, President’s Council of Advisors on Science and Technology; 2015.
13.
Forum on Aging, Disability, and Independence; Board on Health Sciences Policy; Division of Behavioral and Social Sciences and Education; Institute of Medicine; National Research Council.  Hearing Loss and Healthy Aging: Workshop Summary. Washington, DC: The National Academies Press; 2014.
14.
Cassel  C, Penhoet  E, Saunders  R.  Policy solutions for better hearing.  JAMA. 2016;315(6):553-554. doi:10.1001/jama.2016.0044PubMedGoogle ScholarCrossref
15.
Warren  E, Grassley  C.  Over-the-counter hearing aids: the path forward.  JAMA Intern Med. 2017;177(5):609-610. doi:10.1001/jamainternmed.2017.0464PubMedGoogle ScholarCrossref
16.
Huddle  MG, Goman  AM, Kernizan  FC,  et al.  The economic impact of adult hearing loss: a systematic review.  JAMA Otolaryngol Head Neck Surg. 2017;143(10):1040-1048. doi:10.1001/jamaoto.2017.1243PubMedGoogle ScholarCrossref
17.
Simpson  AN, Simpson  KN, Dubno  JR.  Higher health care costs in middle-aged US adults with hearing loss.  JAMA Otolaryngol Head Neck Surg. 2016;142(6):607-609. doi:10.1001/jamaoto.2016.0188PubMedGoogle ScholarCrossref
18.
DeFrank  JT, Bowling  JM, Rimer  BK, Gierisch  JM, Skinner  CS.  Triangulating differential nonresponse by race in a telephone survey.  Prev Chronic Dis. 2007;4(3):A60.PubMedGoogle Scholar
19.
United States Department of Labor. Consumer Price Index. Washington, DC: US Bureau of Labor Statistics; 2016. https://www.bls.gov/cpi/data.htm. Accessed September 20, 2017.
20.
Quan  H, Li  B, Couris  CM,  et al.  Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries.  Am J Epidemiol. 2011;173(6):676-682. doi:10.1093/aje/kwq433PubMedGoogle ScholarCrossref
21.
Kleinman  LC, Norton  EC.  What’s the risk? a simple approach for estimating adjusted risk measures from nonlinear models including logistic regression.  Health Serv Res. 2009;44(1):288-302. doi:10.1111/j.1475-6773.2008.00900.xPubMedGoogle ScholarCrossref
22.
Manning  WG, Mullahy  J.  Estimating log models: to transform or not to transform?  J Health Econ. 2001;20(4):461-494. doi:10.1016/S0167-6296(01)00086-8PubMedGoogle ScholarCrossref
23.
Graubard  BI, Korn  EL.  Predictive margins with survey data.  Biometrics. 1999;55(2):652-659. doi:10.1111/j.0006-341X.1999.00652.xPubMedGoogle ScholarCrossref
24.
Zou  G.  A modified Poisson regression approach to prospective studies with binary data.  Am J Epidemiol. 2004;159(7):702-706. doi:10.1093/aje/kwh090PubMedGoogle ScholarCrossref
25.
Hosmer  DW, Lemeshow  S.  Confidence interval estimation of interaction.  Epidemiology. 1992;3(5):452-456. doi:10.1097/00001648-199209000-00012PubMedGoogle ScholarCrossref
26.
Mohr  PE, Feldman  JJ, Dunbar  JL,  et al.  The societal costs of severe to profound hearing loss in the United States.  Int J Technol Assess Health Care. 2000;16(4):1120-1135. doi:10.1017/S0266462300103162PubMedGoogle ScholarCrossref
27.
Stucky  SR, Wolf  KE, Kuo  T.  The economic effect of age-related hearing loss: national, state, and local estimates, 2002 and 2030.  J Am Geriatr Soc. 2010;58(3):618-619. doi:10.1111/j.1532-5415.2010.02746.xPubMedGoogle ScholarCrossref
28.
Genther  DJ, Frick  KD, Chen  D, Betz  J, Lin  FR.  Association of hearing loss with hospitalization and burden of disease in older adults.  JAMA. 2013;309(22):2322-2324. doi:10.1001/jama.2013.5912PubMedGoogle ScholarCrossref
29.
Deal  JA, Sharrett  AR, Albert  MS,  et al.  Hearing impairment and cognitive decline: a pilot study conducted within the atherosclerosis risk in communities neurocognitive study.  Am J Epidemiol. 2015;181(9):680-690. doi:10.1093/aje/kwu333PubMedGoogle ScholarCrossref
30.
Livingston  G, Sommerlad  A, Orgeta  V,  et al.  Dementia prevention, intervention, and care.  Lancet. 2017;390(10113):2673-2734. doi:10.1016/S0140-6736(17)31363-6PubMedGoogle ScholarCrossref
31.
Strawbridge  WJ, Wallhagen  MI, Shema  SJ, Kaplan  GA.  Negative consequences of hearing impairment in old age: a longitudinal analysis.  Gerontologist. 2000;40(3):320-326. doi:10.1093/geront/40.3.320PubMedGoogle ScholarCrossref
32.
Sudore  RL, Schillinger  D.  Interventions to improve care for patients with limited health literacy.  J Clin Outcomes Manag. 2009;16(1):20-29.PubMedGoogle Scholar
33.
Cohen  JM, Blustein  J, Weinstein  BE,  et al.  Studies of physician-patient communication with older patients: how often is hearing loss considered? a systematic literature review.  J Am Geriatr Soc. 2017;65(8):1642-1649. doi:10.1111/jgs.14860PubMedGoogle ScholarCrossref
34.
Berkman  ND, Sheridan  SL, Donahue  KE, Halpern  DJ, Crotty  K.  Low health literacy and health outcomes: an updated systematic review.  Ann Intern Med. 2011;155(2):97-107. doi:10.7326/0003-4819-155-2-201107190-00005PubMedGoogle ScholarCrossref
35.
Paasche-Orlow  MK, Wolf  MS.  The causal pathways linking health literacy to health outcomes.  Am J Health Behav. 2007;31(1)(suppl 1):S19-S26. doi:10.5993/AJHB.31.s1.4PubMedGoogle ScholarCrossref
Original Investigation
January 2019

Trends in Health Care Costs and Utilization Associated With Untreated Hearing Loss Over 10 Years

Author Affiliations
  • 1Department of Otolaryngology–Head & Neck Surgery, Johns Hopkins School of Medicine, Baltimore, Maryland
  • 2OptumLabs, Cambridge, Massachusetts
  • 3Cochlear Center for Hearing and Public Health, Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland
  • 4Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
  • 5AARP Services, Inc, Washington, DC
  • 6University of California San Francisco School of Nursing, San Francisco
JAMA Otolaryngol Head Neck Surg. 2019;145(1):27-34. doi:10.1001/jamaoto.2018.2875
Key Points

Question  Is untreated hearing loss associated with higher health care costs and utilization?

Findings  In this retrospective, propensity-matched cohort study of claims data, compared with no hearing loss, untreated hearing loss was associated with higher health care costs and a higher risk of 30-day hospital readmission over a 10-year period.

Meaning  Untreated hearing loss may contribute to greater health care costs and utilization.

Abstract

Importance  Nearly 38 million individuals in the United States have untreated hearing loss, which is associated with cognitive and functional decline. National initiatives to address hearing loss are currently under way.

Objective  To determine whether untreated hearing loss is associated with increased health care cost and utilization on the basis of data from a claims database.

Design, Setting, and Participants  Retrospective, propensity-matched cohort study of persons with and without untreated hearing loss based on claims for health services rendered between January 1, 1999, and December 31, 2016, from a large health insurance database. There were 154 414, 44 852, and 4728 participants at the 2-, 5-, and 10-year follow-up periods, respectively. The study was conceptualized and data were analyzed between September 2016 and November 2017.

Exposures  Untreated hearing loss (ie, hearing loss that has not been treated with hearing devices) was identified via claims measures.

Main Outcomes and Measures  Medical costs, inpatient hospitalizations, total days hospitalized, 30-day hospital readmission, emergency department visits, and days with at least 1 outpatient visit.

Results  Among 4728 matched adults (mean age at baseline, 61 years; 2280 women and 2448 men), untreated hearing loss was associated with $22 434 (95% CI, $18 219-$26 648) or 46% higher total health care costs over a 10-year period compared with costs for those without hearing loss. Persons with untreated hearing loss experienced more inpatient stays (incidence rate ratio, 1.47; 95% CI, 1.29-1.68) and were at greater risk for 30-day hospital readmission (relative risk, 1.44; 95% CI, 1.14-1.81) at 10 years postindex. Similar trends were observed at 2- and 5-year time points across measures.

Conclusions and Relevance  Older adults with untreated hearing loss experience higher health care costs and utilization patterns compared with adults without hearing loss. To further define this association, additional research on mediators, such as treatment adherence, and mitigation strategies is needed.

Introduction

More than 38 million adults in the United States experience hearing loss, and the prevalence increases with age such that Quiz Ref IDtwo-thirds of adults older than 70 years have clinically meaningful hearing loss.1 Based on current aging population trends, the number of Americans with hearing loss is projected to increase to more than 73 million by 2060.2Quiz Ref ID Importantly, hearing loss has been associated with negative health outcomes, including cognitive decline,3 incident dementia,4 falls,5 depression,6 reduced quality of life,7 and an increased number of emergency department visits8 and hospitalizations.9 However, fewer than 20% of adults with hearing loss report using hearing aids.10

Hearing loss recently received recognition as a national public health concern.11-15 Reports from the President’s Council of Advisors on Science and Technology12 and the National Academies of Science, Engineering, and Medicine11 have called for policy changes to better understand the influence of hearing loss on health in the United States and to address the condition. These initiatives resulted in the passage of the Over-the-Counter Hearing Aid Act of 2017.15 However, there remains a limited understanding of the broader influence that hearing loss may have on health care cost and resource utilization.8,16,17

This study investigates the association between untreated hearing loss (based on the lack of evidence of hearing device use in claims data and the assumption that relatively few persons with hearing loss use hearing aids10) and health care costs and utilization over a 10-year period using claims data from a large, deidentified administrative claims database from a geographically diverse US health insurance plan.

Methods
Data Source

This retrospective, propensity-matched cohort study used deidentified administrative claims data from the OptumLabs Data Warehouse (OLDW) to identify hearing products purchased or health services rendered between January 1, 1999, and December 31, 2016. Institutional review board approval and patient informed consent were not required because the deidentified data complies with the Health Insurance Portability and Accountability Act Privacy Rule. The OLDW includes deidentified claims data for more than 125 million privately insured and Medicare Advantage (MA) enrollees in a large, private, US health plan from 1993 to the present and represents a diverse population in terms of age, race/ethnicity, and US geographic region. The health plan provides comprehensive insurance coverage for physician, hospital, and prescription drug services, including Part D coverage for MA enrollees. The database includes socioeconomic information, such as race/ethnicity, household income, and education level, for approximately 73% of enrollees. This information was derived from a nationally recognized supplier of consumer marketing data and is a compilation of public data and derived predictive data. While the imputation methods used by this supplier are proprietary, imputation methods for race/ethnicity have been shown in previous studies to have moderate sensitivity (48%), excellent specificity (97%), and moderate positive predictive value (71%) for the purpose of identifying race.18

Outcome Variables

The primary outcome of this study was medical costs, measured as health plan–paid (HPP), out-of-pocket (OOP), and total paid (ie, the sum of all HPP and OOP) costs for all medical services reimbursed by health insurance during the follow-up period. Costs were adjusted to 2015 US dollars using the medical care component of the Consumer Price Index.19 Secondary outcomes included medical costs not related to hearing services and measures of health care utilization, such as the number of inpatient hospitalizations, total days hospitalized, number of readmissions within 30 days of discharge, number of emergency department visits, and number of days with at least 1 outpatient visit (including visits for hearing-related services and visits for non–hearing-related services). Office visit days for hearing loss were defined as the total number of days during follow-up with at least 1 office visit related to hearing services.

Exposure

Older adults with and without hearing loss were identified from OLDW. Eligible participants were 50 years or older. Hearing loss status was determined by the presence of at least 2 claims separated by no more than 730 days with an International Classification of Disease, 9th Revision, Clinical Modification (ICD-9-CM) diagnosis code for hearing loss: V41.2 (problems with hearing), 388.01 (presbycusis), 389 (hearing loss), 389.1x (sensorineural hearing loss, excluding 389.12 [neural hearing loss, bilateral] and 389.14 [central hearing loss]), 389.2x (mixed conductive and sensorineural hearing loss) in any position during the identification period of January 1, 2000, to December 31, 2014. The date of the first claim was designated as the index date. To ensure that complete information was obtained, participants were required to have at least 12 months of continuous enrollment in the plan prior to the index date and at least 2 years of continuous enrollment after the index date. Exclusion criteria included evidence of hearing loss prior to the index date, first evidence of hearing loss experienced during an inpatient stay (indicating an acute issue), evidence of hearing device use via claims data before or within 2 years of the index date, and/or evidence of ototoxic drug use in the 12-month period prior to the index date. Furthermore, individuals were excluded if they had evidence of hearing loss secondary to correctable medical conditions (ie, via surgical measures) or hearing loss fundamentally different from typical peripheral age-related hearing loss (ie, central pathology) including sudden hearing loss (ICD-9-CM code, 388.2), hyperacusis (ICD-9-CM code, 388.42), acoustic nerve disorders (ICD-9-CM code, 388.5), otorrhea (ICD-9-CM code, 388.6x), otalgia (ICD-9-CM code, 388.7x), conductive hearing loss (ICD-9-CM code, 389.0x), neural hearing loss (ICD-9-CM code, 389.12), and/or central hearing loss (ICD-9-CM code, 389.14). A pool of candidate participants with no evidence of hearing loss diagnosis at any point was selected, with the index date set to a random date of service within the capture window. Three nested samples were created, based on data availability for 2-, 5-, and 10-year follow-up periods. The resulting samples are nonmutually exclusive (ie, participants who qualified for the 10-year sample also qualified for the 5- and 2-year samples).

Statistical Analysis

Because of suspected underlying differences in persons with and without hearing loss, matched cohorts were created for each of the 3 follow-up times using a 2-stage process. First, participants with untreated hearing loss were block-matched to persons with no evidence of hearing loss within the cohort year (2, 5, or 10), insurance type (commercial or MA), and cost deciles. Subsequently, nearest-neighbor 1:1 caliper-based greedy propensity score matching without replacement was used, with the following variables measured during the 1 year prior to the index date included in the propensity model: age,2 sex, census geographic region, net worth, race/ethnicity, education level, Charlson Comorbidity Index,20 number of office visits (natural log transformed, both linear and quadratic terms), number of inpatient stays, inpatient length of stay, number of emergency department visits, dementia, mild cognitive impairment, depression, stroke, myocardial infarction, coronary artery disease, breast cancer, prostate cancer, renal cell carcinoma, colorectal cancer, and baseline medical costs (evidence of any costs and total medical costs). Block matching on cost deciles was required owing to the skewed nature of cost, which can result in poorly matched cost distributions. Within each block, caliper matching was used to match persons with untreated hearing loss to the closest person without hearing loss with a propensity score within 0.2 SDs. Unmatched participants were excluded from analysis. The quality of the matched cohorts was assessed using absolute standardized differences (ASD). Following matching, no variable exhibited an ASD greater than 10%, indicating that persons with or without untreated hearing loss were balanced on baseline variables, including follow-up time.

Descriptive statistics were reported for all study measures and baseline covariates. Univariate comparisons were conducted according to the distribution of the data, using t tests, χ2 tests, and Mann-Whitney U tests when appropriate. Statistical significance was assessed at the 2-sided 5% level unless otherwise indicated. Given the large sample size, it was expected a priori that formal statistical tests would show evidence of statistically significant but not clinically significant differences in baseline variables between patients with and without hearing loss even after matching.

Differences in means and proportions and associated 95% CIs are reported. The output of event modeling is also presented as incidence rate ratios, risk ratios, and absolute risk differences.21 In addition to specifications with the cohort indicator only, multivariable analysis was conducted on the matched cohorts to evaluate whether residual confounding remained after matching. Covariates in these models included sociodemographics (eg, age, sex, geographic region, race/ethnicity, net worth), insurance type, and indicators of comorbidity and health-service use measured during the 12-month baseline period. Medical cost variables were analyzed using generalized linear models, assuming a gamma-distributed error and log link.22 This assumption was tested using the modified Park test. The incremental cost of untreated hearing loss was calculated using recycled predictions to facilitate interpretation.23 Negative binomial regression was used to model the number of inpatient hospitalizations and total days hospitalized. The 30-day hospital readmission and emergency department visits were modeled using a modified Poisson regression with robust standard errors and are presented as relative risks (RRs) with 95% CIs.24 Marginal effects, standard errors, and 95% CIs for all multivariable models were calculated using the delta method.25

Analytic files were created using SAS software, version 9.4 (SAS Institute). All analyses were performed using R 3.3.2 (R Core Team, 2016), including R packages ggplot2, data.table, tableone, Rcpp, Zelig, attribrisk, lmtest, sandwich, modmarg, car, ggfortify, and survival; additional R code was custom created. Matching was done in C++ and R; the code is original and was written by one of us (A.D.K.)

Results
Matching

Prior to propensity score matching, there were 77 310 participants with untreated hearing loss and 3 251 863 potential participants without hearing loss. Prior to matching, a greater proportion of persons in the untreated hearing loss cohort were MA enrollees, had greater net worth and education levels, and exhibited a greater level of comorbidity during the 12-month baseline period compared with persons with no evidence of hearing loss. These differences agree with previous research on characteristics of persons with hearing loss and access to hearing care.1 Differences between cohorts were resolved via the matching process. All but 103 persons with hearing loss were matched, and unused data from 3 174 656 unexposed potential participants were dropped. Compared with persons who were included in the study, those persons without hearing loss who were excluded from the study were less likely to be MA enrollees (31.1% vs 17.5% ), were younger (63.8 vs 59.5 years), were less likely to be in the highest net-worth bracket (25.5% vs 19.9%), were less likely to be white (60.4% vs 54.5%), and were less likely to be college educated (15.4%, vs 12.0%). Excluded individuals also had fewer office visits on average (13.4 vs 18.6 visits) and lower average medical costs ($7799 vs $8480) during the baseline year than retained participants had.

Demographic Characteristics

Table 1 lists the characteristics of the matched cohorts at 2-, 5-, and 10-year time points. Between 2000 and 2014, there were 154 414, 44 852, and 4728 participants with and without untreated hearing loss who met study inclusion criteria and had 2, 5, and 10 years of follow-up, respectively. As detailed in Table 1, across the 3 samples of matched persons at 2-, 5-, and 10-year follow-up, the mean age range was 61 to 64 years, 47.6% to 48.6% of participants were women, and 59.1% to 62.9% of participants were white. The percentage of patients with MA insurance ranged from 26.9% to 31.9% (Table 1). Average Charlson comorbidity index scores ranged from 1.12 in the 2-year cohort to 0.68 in the 10-year cohort. Health care utilization was similar across the 3 samples, with the exception that the 10-year cohort had fewer outpatient encounters (mean, 15.55 in the untreated hearing loss group and 14.91 in the no hearing loss group) compared with the 2-year cohort (mean, 18.73 in the untreated hearing loss group and 18.63 in the no hearing loss group) and the 5-year cohort (mean, 17.17 in the untreated hearing loss group and 17.13 in the no hearing loss group).

Health Care Cost

Table 2 details the postmatching unadjusted mean health care cost and resource utilization at 2-, 5-, and 10-year time points, differences between cohorts, and associated 95% CIs. Compared with no hearing loss, Quiz Ref IDuntreated hearing loss was associated with higher medical costs across all 3 cohorts. Unadjusted total cumulative medical costs were 25.9%, 36.9%, and 46.5% higher over the 2-, 5-, and 10-year periods for individuals with untreated hearing loss compared with that of those without hearing loss. This amounted to a mean difference of $3852 (95% CI, $3487-$4217), $11 147 (95% CI, $10 086-$12 208), and $22 434 (95% CI, $18 219-$26 648) higher total health care costs for the individuals with untreated hearing loss compared with health care costs for those without hearing loss (Figure 1 and Table 3). Unadjusted OOP costs were 19.1%, 23.3%, and 34.0% higher over the 2-, 5-, and 10-year periods, respectively, for patients with untreated hearing loss than for patients with no hearing loss. Notably, among the untreated hearing loss cohort, mean costs specifically related to hearing loss were relatively minor within health care expenditures of $382, $470, and $556 at 2, 5, and 10 years, respectively. Results from adjusted multivariable models were substantively unchanged.

Health Resource Utilization

Quiz Ref IDParticipants with untreated hearing loss experienced significantly more inpatient stays compared with participants without hearing loss at 2 years (incident rate ratio [IRR], 1.20; 95% CI, 1.15-1.25), 5 years (IRR, 1.30; 95% CI, 1.23-1.38), and 10 years (IRR, 1.47; 95% CI,1.29-1.68) following the index date. Moreover, participants with untreated hearing loss spent more days in the hospital on average with 0.26 (95% CI, 0.20-0.33), 0.75 (95% CI, 0.58-0.92), and 2.10 (95% CI, 1.36-2.84) more days spent in the hospital over the 2-, 5-, and 10-year periods, respectively (Table 2). Similar results were found for outpatient visits. Over the 2 years following the index date, participants with untreated hearing loss had an average of 7.5 (95% CI, 7.1-7.8) more outpatient visit days. By 10 years, patients with untreated hearing loss had 52.2 (95% CI, 45.9-58.5) more outpatient visit days. Participants with untreated hearing loss had a 16.9% increased risk of an emergency department visit within 2 years (RR, 1.17; 95% CI, 1.15-1.19), a 16.8% increased risk over 5 years (RR, 1.17; 95% CI, 1.15-1.19), and a 17.0% increased risk over 10 years (RR, 1.17; 95% CI, 1.12-1.22). Figure 2 displays the unadjusted difference in risk of 30-day readmissions. Participants with untreated hearing loss had a 29% increased risk of a 30-day hospital readmission within 2 years (RR, 1.29; 95% CI, 1.19-1.39), a 28% increase in risk over 5 years (RR, 1.28; 95% CI, 1.16-1.42), and a 44% increase in risk over 10 years (RR, 1.44; 95% CI, 1.14-1.81). Results of the adjusted multivariable analysis were similar to those of the unadjusted analyses.

Discussion

In a matched cohort study of up to 154 414 individuals in a large administrative claims database, untreated hearing loss was associated with higher health care costs, increased risk of 30-day hospital readmission, more inpatient and outpatient stays, increased risk of emergency department visits, and longer length of hospital stays at 2, 5, and 10 years following initial hearing loss diagnosis. Over a 10-year period, persons with untreated hearing loss incurred an average $22 434 more in health care costs than persons without evidence of hearing loss. Importantly, the magnitude of most of these associations appears to compound over time. These findings suggest that persons with untreated hearing loss experience significantly higher health care costs and higher health care utilization rates than those without hearing loss. To our knowledge, this is the first 10-year analysis of the association of untreated hearing loss with health care cost and utilization measures using real-world claims data.

Research on health care costs related to hearing loss is sparse and varies because studies have often examined the overall societal economic influence of hearing loss on projected medical expenses and lost labor productivity rather than on the direct and indirect health care costs that were actually incurred.16,26 On the basis of a simulated model from California Medicaid data, medical costs directly related to hearing loss diagnosis and treatment 1 year after diagnosis were $1292 per person older than 65 years with any type of hearing loss (ie, mild to functionally deaf).27 Furthermore, according to data from the 2000 to 2010 Medical Expenditure Panel Surveys (pooled), which includes self-reported hearing loss, persons older than 65 years with hearing loss had, on average, an estimated $392 (95% CI, $277-$392) in excess medical expenditures.8 These simulated results are relatively low compared with the present study’s results.

In the only other, to our knowledge, claims-based study examining hearing loss and medical expenditures, health care costs in patients with hearing loss were substantially higher than those of patients without hearing loss.17 Researchers conducted a matched analysis of 561 764 individuals with and without hearing loss claims (based on ICD-9-CM codes V41.2, V72.1x, 388.00, 388.01, 388.40, 388.43, 388.44, 388.5, 389.1x, and 389.2x) over an 18-month period using the Truven Health MarketScan database. They found that those with hearing loss spent more, on average, in health care costs ($14 165; 95% CI, $14 091-$14 239) than did persons without hearing loss ($10 629; 95% CI, $10 576-$10 681). The difference of $3536 over a 1.5-year period is relatively similar to the difference of $4764 over the 2-year period observed in the present study.

The association between hearing loss and health care service utilization remains mostly unexplored. Two studies found that hearing loss was associated with an increased risk of hospitalization based on objective audiometric measures and self-reported hospitalization rates.9,28 In 1 study, individuals with hearing loss had a 16% (HR, 1.16; 95% CI, 1.04-1.29) and 21% (HR, 1.21; 95% CI, 1.06-1.38) greater risk of hospitalization compared with those without hearing loss.9 In addition, the previously noted study of Medical Expenditure Panel Survey data reported increased odds of emergency department visits associated with hearing loss.8 Our results are broadly consistent with those of these earlier studies and further suggest that hearing loss is associated with increased average length of hospital stay and risk of 30-day readmissions.

The observed associations between hearing loss and health care utilization and costs is possibly explained by the potential influence of hearing loss on cognitive (eg, particularly dementia),3,4,29,30 physical (eg, falls),5 and psychosocial6,31 function. The association between hearing loss and health outcomes may be further mediated by the effects of hearing loss on patient-clinician communication.32,33 Poor communication has previously been associated with poorer health care outcomes, poor treatment adherence, lower patient satisfaction, and increased health care costs.34,35 This study supports the need for future research to understand the role of hearing loss on patient-clinician communication and the potential influence of hearing care, including devices and services, in mitigating the overall association between hearing loss and health care cost. Importantly, given the minimal risk involved and the potential reward, health care clinics and/or institutions could consider awareness and communication training for clinicians and screening programs to identify and address hearing loss as a potential communication barrier.

Limitations

This study used a large, deidentified, geographically diverse database of administrative claims that includes health encounter information for more than 125 million commercial and MA insurance enrollees in the United States. There are several limitations inherent to claims data, including its reliance on coding processes designed for billing purposes rather than research. Capturing the exposure (ie, untreated hearing loss) in a population is likely limited to individuals who feel that hearing loss is a great burden on their lifestyle, those with the means to seek and continuously maintain health care, and those with coverage for hearing services. Quiz Ref IDBecause of limitations in the coding and reimbursement for hearing care and hearing devices under some health plan policies, and because some individuals do not seek care for hearing loss, the group without hearing loss diagnosis may include individuals with uncoded hearing loss. However, such a bias would likely lead to underestimation of the differences between these cohorts. Furthermore, hearing device use as a covariate in these data would be unreliable given that many sales occur outside the medical claims data and that access is highly related to socioeconomic status; thus both cohorts may include individuals with treated hearing loss. Again, this resultant bias may underestimate the differences between cohorts. Finally, claims data do not account for indirect costs of untreated hearing loss, such as lost productivity and wages.

Other limitations not related to the data source exist. Though examination of ASDs suggests well-matched cohorts, after matching, a few statistically significant differences were still present between cohorts. However, these differences are to be expected based on the large sample size, and likely do not represent clinically meaningful differences between cohorts. As with most analyses of observational data, despite careful matching, it is possible that residual unmeasured confounding exists owing to a lack of randomized treatment allocation, prohibiting causal inferences from these data. Furthermore, findings may be affected by detection bias because individuals with hearing loss may have frequent interaction with the health care system. Finally, survival bias in the 10-year cohort and a lack of Medicaid and uninsured participants in the database may limit generalizability.

Conclusions

The present study suggests that hearing loss is associated with increased health care expenditures and resource utilization. Notably, hearing loss was associated with an average 46.5% increase in health care costs and a 44% increase in risk of 30-day readmission over a 10-year period. Awareness of the burden that hearing loss places on individuals, insurers, and hospitals contributes to the growing evidence of hearing loss as a public health concern.

Back to top
Article Information

Accepted for Publication: August 26, 2018.

Corresponding Author: Nicholas S. Reed, AuD, Johns Hopkins University, Cochlear Center for Hearing and Public Health, 2024 E Monument Street, Baltimore, MD 21205 (nreed9@jhmi.edu).

Published Online: November 8, 2018. doi:10.1001/jamaoto.2018.2875

Author Contributions: Drs Altan and Kravetz had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Reed, Yeh, Kravetz, Wallhagen, Lin.

Acquisition, analysis, or interpretation of data: Reed, Altan, Deal, Yeh, Kravetz, Wallhagen.

Drafting of the manuscript: Reed, Altan, Kravetz, Wallhagen.

Critical revision of the manuscript for important intellectual content: Reed, Altan, Deal, Yeh, Wallhagen, Lin.

Statistical analysis: Altan, Kravetz.

Administrative, technical, or material support: Altan, Yeh, Lin.

Study supervision: Reed, Altan, Yeh, Lin.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Reed reports grants from NIH KL2TR001077 during the conduct of the study and Scientific Advisory Board Member (nonfinancial relationship) to Clearwater Clinical. Dr Altan is an employee of OptumLabs. The AARP is a research partner and founding consumer advocate in OptumLabs and as a partner they fund 2 full-time equivalents to work on various projects in health care policy for the aging. Dr Altan’s work on this project was a part of that 2 full-time equivalent allotment. However, employment with OptumLabs was not dependent on that funding and the funding support work on multiple projects being conducted at OptumLabs. Dr Yeh reports a vendor client relationship with Hear USA that provides hearing services to AARP members. Dr Wallhagen is on the Board of the Hearing Loss Association of America in a noncompensated position and receives no payments and pay for all travel and other expenses. Dr Lin reports grants from National Institutes of Health and grants from Eleanor Schwartz Charitable Foundation during the conduct of the study; personal fees from Cochlear Ltd, personal fees from Amplifon, and personal fees from Boehringer Ingelheim outside the submitted work. No other disclosures were reported.

Funding/Support: This work was supported by AARP and AARP Services, Inc. Dr Reed was supported by NIH grant NIH KL2TR001077. Dr Deal was supported by NIH/NIA grant K01AG054693. Dr Lin was supported by R01AG055426, R01HL096812, R33DC015062, and the Eleanor Schwartz Charitable Foundation. Drs Deal, Reed, and Lin were supported in part by the Cochlear Center for Hearing and Public Health.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: The authors thank Tanya Natwick, BA, (OptumLabs) for her assistance in final preparation of this manuscript. Ms Natwick received no additional compensation beyond that of the regular course of her employment for her assistance with this article.

References
1.
Lin  FR, Niparko  JK, Ferrucci  L.  Hearing loss prevalence in the United States.  Arch Intern Med. 2011;171(20):1851-1852. doi:10.1001/archinternmed.2011.506PubMedGoogle ScholarCrossref
2.
Goman  AM, Reed  NS, Lin  FR.  Addressing estimated hearing loss in adults in 2060.  JAMA Otolaryngol Head Neck Surg. 2017;143(7):733-734. doi:10.1001/jamaoto.2016.4642PubMedGoogle ScholarCrossref
3.
Lin  FR, Yaffe  K, Xia  J,  et al; Health ABC Study Group.  Hearing loss and cognitive decline in older adults.  JAMA Intern Med. 2013;173(4):293-299. doi:10.1001/jamainternmed.2013.1868PubMedGoogle ScholarCrossref
4.
Lin  FR, Metter  EJ, O’Brien  RJ, Resnick  SM, Zonderman  AB, Ferrucci  L.  Hearing loss and incident dementia.  Arch Neurol. 2011;68(2):214-220. doi:10.1001/archneurol.2010.362PubMedGoogle ScholarCrossref
5.
Lin  FR, Ferrucci  L.  Hearing loss and falls among older adults in the United States.  Arch Intern Med. 2012;172(4):369-371. doi:10.1001/archinternmed.2011.728PubMedGoogle ScholarCrossref
6.
Wallhagen  MI, Strawbridge  WJ, Shema  SJ, Kurata  J, Kaplan  GA.  Comparative impact of hearing and vision impairment on subsequent functioning.  J Am Geriatr Soc. 2001;49(8):1086-1092. doi:10.1046/j.1532-5415.2001.49213.xPubMedGoogle ScholarCrossref
7.
Dalton  DS, Cruickshanks  KJ, Klein  BE, Klein  R, Wiley  TL, Nondahl  DM.  The impact of hearing loss on quality of life in older adults.  Gerontologist. 2003;43(5):661-668. doi:10.1093/geront/43.5.661PubMedGoogle ScholarCrossref
8.
Foley  DM, Frick  KD, Lin  FR.  Association between hearing loss and healthcare expenditures in older adults.  J Am Geriatr Soc. 2014;62(6):1188-1189. doi:10.1111/jgs.12864PubMedGoogle ScholarCrossref
9.
Genther  DJ, Betz  J, Pratt  S,  et al; Health, Aging and Body Composition Study.  Association between hearing impairment and risk of hospitalization in older adults.  J Am Geriatr Soc. 2015;63(6):1146-1152. doi:10.1111/jgs.13456PubMedGoogle ScholarCrossref
10.
Chien  W, Lin  FR.  Prevalence of hearing aid use among older adults in the United States.  Arch Intern Med. 2012;172(3):292-293. doi:10.1001/archinternmed.2011.1408PubMedGoogle ScholarCrossref
11.
National Academies of Sciences, Engineering, and Medicine.  Hearing Health Care for Adults: Priorities for Improving Access and Affordability. Washington, DC: The National Academies Press; 2016.
12.
President’s Council of Advisors on Science and Technology (US).  Aging America and Hearing Loss: Imperative of Improved Hearing Technologies. Washington, DC: Executive Office of the President, President’s Council of Advisors on Science and Technology; 2015.
13.
Forum on Aging, Disability, and Independence; Board on Health Sciences Policy; Division of Behavioral and Social Sciences and Education; Institute of Medicine; National Research Council.  Hearing Loss and Healthy Aging: Workshop Summary. Washington, DC: The National Academies Press; 2014.
14.
Cassel  C, Penhoet  E, Saunders  R.  Policy solutions for better hearing.  JAMA. 2016;315(6):553-554. doi:10.1001/jama.2016.0044PubMedGoogle ScholarCrossref
15.
Warren  E, Grassley  C.  Over-the-counter hearing aids: the path forward.  JAMA Intern Med. 2017;177(5):609-610. doi:10.1001/jamainternmed.2017.0464PubMedGoogle ScholarCrossref
16.
Huddle  MG, Goman  AM, Kernizan  FC,  et al.  The economic impact of adult hearing loss: a systematic review.  JAMA Otolaryngol Head Neck Surg. 2017;143(10):1040-1048. doi:10.1001/jamaoto.2017.1243PubMedGoogle ScholarCrossref
17.
Simpson  AN, Simpson  KN, Dubno  JR.  Higher health care costs in middle-aged US adults with hearing loss.  JAMA Otolaryngol Head Neck Surg. 2016;142(6):607-609. doi:10.1001/jamaoto.2016.0188PubMedGoogle ScholarCrossref
18.
DeFrank  JT, Bowling  JM, Rimer  BK, Gierisch  JM, Skinner  CS.  Triangulating differential nonresponse by race in a telephone survey.  Prev Chronic Dis. 2007;4(3):A60.PubMedGoogle Scholar
19.
United States Department of Labor. Consumer Price Index. Washington, DC: US Bureau of Labor Statistics; 2016. https://www.bls.gov/cpi/data.htm. Accessed September 20, 2017.
20.
Quan  H, Li  B, Couris  CM,  et al.  Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries.  Am J Epidemiol. 2011;173(6):676-682. doi:10.1093/aje/kwq433PubMedGoogle ScholarCrossref
21.
Kleinman  LC, Norton  EC.  What’s the risk? a simple approach for estimating adjusted risk measures from nonlinear models including logistic regression.  Health Serv Res. 2009;44(1):288-302. doi:10.1111/j.1475-6773.2008.00900.xPubMedGoogle ScholarCrossref
22.
Manning  WG, Mullahy  J.  Estimating log models: to transform or not to transform?  J Health Econ. 2001;20(4):461-494. doi:10.1016/S0167-6296(01)00086-8PubMedGoogle ScholarCrossref
23.
Graubard  BI, Korn  EL.  Predictive margins with survey data.  Biometrics. 1999;55(2):652-659. doi:10.1111/j.0006-341X.1999.00652.xPubMedGoogle ScholarCrossref
24.
Zou  G.  A modified Poisson regression approach to prospective studies with binary data.  Am J Epidemiol. 2004;159(7):702-706. doi:10.1093/aje/kwh090PubMedGoogle ScholarCrossref
25.
Hosmer  DW, Lemeshow  S.  Confidence interval estimation of interaction.  Epidemiology. 1992;3(5):452-456. doi:10.1097/00001648-199209000-00012PubMedGoogle ScholarCrossref
26.
Mohr  PE, Feldman  JJ, Dunbar  JL,  et al.  The societal costs of severe to profound hearing loss in the United States.  Int J Technol Assess Health Care. 2000;16(4):1120-1135. doi:10.1017/S0266462300103162PubMedGoogle ScholarCrossref
27.
Stucky  SR, Wolf  KE, Kuo  T.  The economic effect of age-related hearing loss: national, state, and local estimates, 2002 and 2030.  J Am Geriatr Soc. 2010;58(3):618-619. doi:10.1111/j.1532-5415.2010.02746.xPubMedGoogle ScholarCrossref
28.
Genther  DJ, Frick  KD, Chen  D, Betz  J, Lin  FR.  Association of hearing loss with hospitalization and burden of disease in older adults.  JAMA. 2013;309(22):2322-2324. doi:10.1001/jama.2013.5912PubMedGoogle ScholarCrossref
29.
Deal  JA, Sharrett  AR, Albert  MS,  et al.  Hearing impairment and cognitive decline: a pilot study conducted within the atherosclerosis risk in communities neurocognitive study.  Am J Epidemiol. 2015;181(9):680-690. doi:10.1093/aje/kwu333PubMedGoogle ScholarCrossref
30.
Livingston  G, Sommerlad  A, Orgeta  V,  et al.  Dementia prevention, intervention, and care.  Lancet. 2017;390(10113):2673-2734. doi:10.1016/S0140-6736(17)31363-6PubMedGoogle ScholarCrossref
31.
Strawbridge  WJ, Wallhagen  MI, Shema  SJ, Kaplan  GA.  Negative consequences of hearing impairment in old age: a longitudinal analysis.  Gerontologist. 2000;40(3):320-326. doi:10.1093/geront/40.3.320PubMedGoogle ScholarCrossref
32.
Sudore  RL, Schillinger  D.  Interventions to improve care for patients with limited health literacy.  J Clin Outcomes Manag. 2009;16(1):20-29.PubMedGoogle Scholar
33.
Cohen  JM, Blustein  J, Weinstein  BE,  et al.  Studies of physician-patient communication with older patients: how often is hearing loss considered? a systematic literature review.  J Am Geriatr Soc. 2017;65(8):1642-1649. doi:10.1111/jgs.14860PubMedGoogle ScholarCrossref
34.
Berkman  ND, Sheridan  SL, Donahue  KE, Halpern  DJ, Crotty  K.  Low health literacy and health outcomes: an updated systematic review.  Ann Intern Med. 2011;155(2):97-107. doi:10.7326/0003-4819-155-2-201107190-00005PubMedGoogle ScholarCrossref
35.
Paasche-Orlow  MK, Wolf  MS.  The causal pathways linking health literacy to health outcomes.  Am J Health Behav. 2007;31(1)(suppl 1):S19-S26. doi:10.5993/AJHB.31.s1.4PubMedGoogle ScholarCrossref
×