Observational Study Medicine ®

OPEN

Increased risk of depression in patients with

acquired sensory hearing loss
A 12-year follow-up study
Wei-Ting Hsu, MDa, Chih-Chao Hsu, MDb, Ming-Hsun Wen, MDc, Hong-Ching Lin, MDa,d,
Hsun-Tien Tsai, MD, PhDa, Peijen Su, MD, PhDd,e, ∗Chi-Te Sun, MScf, Cheng-Li
∗
Lin, MScg,
Chung-Yi Hsu, MD, PhDh, Kuang-Hsi Chang, PhDi, , Yi-Chao Hsu, PhDj,

Abstract
Acquired sensory hearing loss (SHL) is suggested to be associated with depression. However, some studies have reported
conflicting results. Our study investigated the relationship between the prevalence of SHL and the incidence of depression over
12 years of follow-up by using data from the Taiwan National Health Insurance Research Database (NHIRD). We sought to determine
the association between SHL and subsequent development of depression and discuss the pathophysiological mechanism
underlying the association.
Patients with SHL were identified from the NHIRD (SHL cohort). A non-SHL cohort, comprising patients without SHL frequency-
matched with the SHL patients according to age group, sex, and the year of diagnosis of SHL at the ratio of 1:4, was constructed,
and the incidence of depression was evaluated in both cohorts. A multivariable model was adjusted for age, sex, and comorbidity.
The SHL cohort and non-SHL cohort comprised 5043 patients with SHL and 20,172 patients without SHL, respectively. The
incidences density rates were 9.50 and 4.78 per 1000 person-years in the SHL cohort and non-SHL cohort, respectively. After
adjustment for age, sex, and comorbidities, the risk of depression was higher in the SHL cohort than in the non-SHL cohort (hazard
ratio = 1.73, 95% confidence interval = 1.49–2.00).
Acquired SHL may increase the risk of subsequent depression. The results demonstrated that SHL was an independent risk factor
regardless of sex, age, and comorbidities. Moreover, a strong association between hearing loss and subsequent depression among
Taiwanese adults of all ages, particularly those aged 49 and >65 years and without using steroids for the treatment of SHL was
observed. Prospective clinical and biomedical studies on the relationship between hearing loss and depression are warranted for
determining the etiopathology.
Abbreviations: 5-HT = 5-hydroxytryptamine, ADLs = activities of daily living, CAD = coronary artery disease, CI = confidence
intervals, CKD = chronic kidney disease, COPD = chronic obstructive pulmonary disease, HR = hazard ratios, LHID200 =
Longitudinal Health Insurance Database of 2000, NHI = National Health Insurance, NHIRD = National Health Insurance Research
Database, SHL = Sensory Hearing Loss, SSRIs = selective serotonin reuptake inhibitors, US = United States.
Keywords: depression, NHIRD, sensory hearing loss

Editor: Giovanni Tarantino.

W-TH and C-CH contributed equally to this work.
Author contributions: conception/design: Y-CH; provision of study materials: K-HC, Y-CH; collection and/or assembly of data: K-HC, Y-CH; data analysis and
interpretation: all authors; manuscript writing: W-TH, C-CH, Y-CH; final approval of manuscript: all authors.
This study is supported in part by Taiwan Ministry of Health and Welfare Clinical Trial and Research Center of Excellence (MOHW104-TDU-B-212-113002); China
Medical University Hospital, Academia Sinica Taiwan Biobank, Stroke Biosignature Project (BM104010092); NRPB Stroke Clinical Trial Consortium (MOST 103-2325-B-
039-006); Tseng-Lien Lin Foundation, Taichung, Taiwan; Taiwan Brain Disease Foundation, Taipei, Taiwan; Katsuzo and Kiyo Aoshima Memorial Funds, Japan; and
CMU under the Aim for Top University Plan of the Ministry of Education, Taiwan. This study is also supported from the financial support by grants from Ministry of
Science and Technology (MOST103-2314-B-715-001-MY2, and MOST104-2314-B-715-003-MY3), and Mackay Medical College (RD1050179). The funders had no
role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. No additional external funding was received for this study.
The authors have no conflicts of interest to disclose.
a
Department of Psychiatry, Kaohsiung Veterans General Hospital, Kaohsiung, b Department of Otorhinolaryngology and Head and Neck Surgery, Mackay Memorial
Hospital, Taipei, c Department of Otorhinolaryngology and Head and Neck Surgery, Far Eastern Memorial Hospital, New Taipei City, d Department of Audiology and
Speech-Language Pathology, Mackay Medical College, e Department of Family Medicine, Mackay Memorial Hospital, Taipei, f Fu Jen Catholic University Graduate
Institution of Business Administration, g Management Office for Health Data, China Medical University Hospital, h Graduate Institute of Clinical Medical Sciences, Center
College of Medicine, i Department of Public Health, China Medical University, Taichung, j Institute of Biomedical Sciences, Mackay Medical College, Taipei, Taiwan.
∗
Correspondence: Yi-Chao Hsu, Institute of Biomedical Sciences, Mackay Medical College, No. 46, Sec. 3, Zhongzheng Rd., Sanzhi Dist., New Taipei City 252, Taiwan
(e-mail: hsuyc@mmc.edu.tw); Kuang-Hsi Chang, Department of Public Health, China Medical University, Taichung, Taiwan (e-mail: s76881072@yahoo.com.tw).
Copyright © 2016 the Author(s). Published by Wolters Kluwer Health, Inc. All rights reserved.
This is an open access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is
permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the
journal.
Medicine (2016) 95:44(e5312)
Received: 15 January 2016 / Received in final form: 11 October 2016 / Accepted: 12 October 2016
http://dx.doi.org/10.1097/MD.0000000000005312

1
Hsu et al. Medicine (2016) 95:44 Medicine

1. Introduction 2.3. Comorbidity

Hearing impairment is a common ailment associated with ageing The following baseline comorbidities were considered covariates:
and the most common cause of disability worldwide. Approxi- cirrhosis (ICD-9-CM code 571), rheumatoid arthritis (ICD-9-
mately 1 in 4 adults aged ≥45 years experiences mild or greater CM code 714), hypertension (ICD-9-CM codes 401-405),
hearing loss.[1] World Health Organization estimated 360 million hyperlipidemia (ICD-9-CM code 272), diabetes mellitus (ICD-
persons, about 5.3% of the world’s population, suffered from 9-CM code 250), asthma (ICD-9-CM code 493), chronic kidney
disabling hearing loss.[2] Depression is a common mental disease (CKD, ICD-9-CM codes 580-589), coronary artery
disorder, which affect 350 million people in the world.[3] disease (CAD, ICD-9-CM codes 410-414), alcohol-related illness
Unipolar depressive disorders and adult-onset hearing loss, the (ICD-9-CM codes 291, 303, 305, 571.0, 571.1, 571.2, 571.3,
most common neuropsychiatric conditions, and sense organ 790.3, A215, and V11.3), anxiety (ICD-9-CM code 300.00),
disorder, respectively, are the first and second leading nonfatal chronic obstructive pulmonary disease (COPD, ICD-9-CM code
causes of year loss due to disability among adults in high-income 491, 492, 496), stroke (ICD-9-CM code 430-438). Steroids was
countries.[1] Hearing loss may increase deterioration of health- known for treatment of SHL.[15] The ICD-9-CM code we used for
related quality of life, and hearing impairment influences social anxiety was “unspecified anxiety disorder” rather than other
behavior, making affected people prone to depression, anxiety, specific diagnosis of anxiety disorder, such as generalized anxiety
interpersonal sensitivity, and hostility.[4–6] Several cross-sectional disorder or panic disorder.
studies have reported that hearing loss is independently
associated with depression in the elderly population.[6–8] 2.4. Statistical analysis
However, some studies have reported conflicting results.[9,10]
A nationwide study reported a strong association between The x2 test and t test were used for analyzing the differences
hearing impairment and depression among adults of all ages in between the cohorts in categorical and continuous variables,
the United States.[11] In this 12-year nationwide population- respectively. The cumulative incidence of depression between the
based cohort study, we explored the relationship between 2 cohorts was plotted using the Kaplan–Meier method, and the
acquired sensory hearing loss (SHL) and the incidence of difference was analyzed using a log-rank test. The incidence
depression by comparing Taiwanese patients with and without density rates of depression in both cohorts were calculated.
hearing impairment. Univariable and multivariable Cox proportional hazard regres-
sion analyses were conducted for estimating the relative hazard
ratios (HRs) and 95% confidence intervals (CIs) of depression in
2. Methods the SHL cohort compared with the non-SHL cohort. The
multivariable model was adjusted for age, sex, and comorbidity
2.1. Data source
variables that had a significant difference according to Table 1.
Data from the Longitudinal Health Insurance Database of 2000 All data analyses were conducted using the SAS statistical
(LHID2000) released by the National Health Research Institutes package (version 9.3 for Windows; SAS institute Inc, Cary, NC).
of Taiwan were used. Briefly, the Taiwan National Health A 2-tailed P value <0.05 indicated statistical significance.
Insurance (NHI) program is a universal healthcare system that
covers 99% of the country’s population of 23 million.[12] The
3. Results
details of this program have been described previously.[13,14] The
LHID2000 contains detailed records of each visit of each patient, Table 1 presents the demographic characteristics and comorbid-
including outpatient visits, emergency department visits, and ities of the 5043 patients in the SHL cohort and 20,172 patients in
hospital admission. The LHID2000 also includes principal and the non-SHL cohort. Most patients in both cohorts were aged
secondary diagnostic codes, prescription orders, and claimed ≥65 years (50%) and men (61.9%). The mean ages of the patients
expenses. The diagnoses and procedures are coded according to in the SHL cohort and non-SHL cohort were 61.8 ± 17.0 years
the International Classification of Diseases, Ninth Revision, and 61.2 ± 17.0 years, respectively. Comorbidities, except for
Clinical Modification (ICD-9-CM). The study protocol was rheumatoid arthritis, were more prevalent in the SHL cohort than
approved by the Ethics Review Board of China Medical in the non-SHL cohort (P <0.05). The mean follow-up periods
University (CMUH-104-REC2-115). for the SHL cohort and non-SHL cohort were 5.70 ± 3.14 years
and 5.54 ± 3.15 years, respectively. After 12 years of follow-up,
the cumulative incidence of depression was higher in the SHL
2.2. Sampled patients
cohort than in the non-SHL cohort (P <0.001; Fig. 1).
Patients aged >20 years from 2000 to 2011 and newly diagnosed The incidences density rates in the SHL cohort and non-SHL
with SHL (ICD-9 codes 388.01 and 389.10-389.12) were cohort were 9.50 (crude HR = 1.99, 95% CI = 1.72–2.31) and
identified (SHL cohort). The index date was the date of initial 4.78 per 1000 person-years, respectively (Table 2). After
diagnosis of SHL. Patients without SHL were randomly selected adjustment for age, sex, and comorbidities, the risk of depression
from the LHID2000 for inclusion in a non-SHL cohort and were was higher in the SHL cohort than in the non-SHL cohort (aHR =
frequency-matched with the SHL patients according to age group 1.73, 95% CI = 1.49–2.00). The incidence of depression
(every 5-year span), sex, and the year of SHL diagnosis at the ratio increased with age and was higher in women than in men.
of 1:4. The index date for the non-SHL patients was randomly The multivariable analysis revealed that the risk of depression
assigned as a day and month in the index year of the matched SHL was 1.35-fold higher in women than in man (95% CI =
patient. Both cohorts excluded patients with a history of depression 1.17–1.56). The risk of depression was higher in patients with
(ICD-9-CM codes 296.2, 296.3, 300.4, and 311) at the baseline the comorbidities of CAD (aHR = 1.52, 95% CI = 1.28–1.81),
and those with incomplete information on age or sex. All patients alcohol-related illness (aHR = 1.61, 95% CI = 1.12–2.29), anxi-
were followed from the index date until the diagnosis of ety (aHR = 2.38, 95% CI = 1.95–2.89), stroke (aHR = 1.31, 95%
depression, date of NHI withdrawal, or end of 2011. CI = 1.01–1.70).

2
Hsu et al. Medicine (2016) 95:44 www.md-journal.com

Table 1
Demographic characteristics and comorbidities in cohorts with and without sensory hearing loss.
Sensory hearing loss
No Yes
Variable N = 20,172 N = 5043 P value
Age, y 0.99
34 1796 (8.90) 449 (8.90)
35–49 3132 (15.5) 783 (15.5)
50–64 5168 (25.6) 1292 (25.6)
65+ 10,076 (50.0) 2519 (50.0)
∗
Mean ± SD 61.2 (17.0) 61.8 (17.0) 0.03
Sex 0.99
Women 7688 (38.1) 1922 (38.1)
Men 12,484 (61.9) 3121 (61.9)
Comorbidity
Cirrhosis 3613 (17.9) 1275 (25.3) <0.001
Rheumatoid arthritis 44 (0.22) 17 (0.34) 0.12
Hypertension 8726 (43.3) 2520 (50.0) <0.001
Hyperlipidemia 4379 (21.7) 1442 (28.6) <0.001
Diabetes mellitus 2515 (12.5) 692 (13.7) 0.02
Asthma 1731 (8.58) 555 (11.0) <0.001
Chronic kidney disease 438 (2.17) 157 (3.11) <0.001
Coronary artery disease 4345 (21.5) 1504 (29.8) <0.001
Alcohol-related illness 564 (2.80) 183 (3.63) 0.002
Anxiety 1098 (5.44) 557 (11.1) <0.001
COPD 3308 (16.4) 1218 (24.2) <0.001
Stroke 284 (5.63) 1366 (6.77) 0.003
Medication
Steroid 2209 (11.0) 732 (14.5) <0.001
x test.
2
∗
T test.

The age-specific relative risk of depression for the SHL cohort Despite a strong association between depression and SHL, the
compared with the non-SHL cohort was significant in all age etiologic mechanisms underlying the relationship between
groups, except in the age group of 50 to 64 years (Table 3). The hearing loss and depression have not been well established.
sex-specific relative risk of depression for the SHL cohort Hearing impairment elicits a strong feeling of isolation and
compared with the non-SHL cohort was significant in both frustration in patients, particularly when they fail to communi-
women (aHR = 1.55, 95% CI = 1.24–1.95) and men (aHR = cate with others.[16,17] It can also impose a heavy social and
1.88, 95% CI = 1.54–2.29). After stratification for comorbidity, economic burden on the patient and family.[18] Patients with
the relative risk of depression was higher in the SHL cohort than moderate to severe hearing loss were more likely to have impaired
in the non-SHL cohort for patients without comorbidity (aHR = activities of daily living (ADLs) and instrumental ADLs.[19] The
1.83, 95% CI = 1.31–2.55) and those with comorbidity (aHR = declined quality of life and social isolation may lead to
1.87, 95% CI = 1.58–2.20). depression. On the other hand, serotonin (5-hydroxytryptamine,
Steroids were known for treatment of SHL.[15] In order to 5-HT) is a neurotransmitter associated with depression. Selective
investigate whether SHL patients treated with steroids could serotonin reuptake inhibitors (SSRIs) are used for treating
lower the risk of subsequent depression, we further added the depression by modulating the serotonin pathway.[20,21] An
analysis of steroids treatment of SHL patients. As shown in animal study reported that sertraline, an SSRI used widely in
Table 3, SHL patients without steroids treatment significantly depression, has a protective effect on cisplatin ototoxicity, which
increase the risk of subsequent depression (aHR = 1.78, 95% leads to SHL.[22] Another study on rats demonstrated that early
CI = 1.51–2.08). The relative risk of depression was significantly hearing loss affects the ability of 5-HT receptor activation to
higher in the SHL cohort without steroids treatment (aHR = 1.78, modulate primary auditory cortex excitability.[23] Acoustic
95% CI = 1.51–2.08) than in the SHL cohort with steroids trauma in an animal model induced substantial hearing loss
treatment (aHR = 1.40, 95% CI = 0.94–2.09). Taken together, and caused selective upregulation of serotonin receptor genes in
our study demonstrated no increased risk of subsequent the inferior colliculus.[24] These 3 studies have demonstrated that
depression among patients with steroids treatment. On the other hearing loss may induce plasticity in the excitatory and inhibitory
hand, those without steroids treatment were noted having neurotransmitter systems in the central auditory brain regions.
increased risk of subsequent depression. However, information on networks and etiologic mechanisms
between auditory damage and depression via the serotonin
pathway is scant. Clinical and experimental studies should clarify
4. Discussion
these networks and mechanisms in the future.
This is a population-based study in Taiwan to investigate SHL Risk factors for hearing loss include race, age, the male sex,
as a risk factor for depression by using a matched cohort and a diabetes mellitus, exposure to noise, and heavy smoking.[25] A
12-year follow-up period. Our data demonstrated that SHL may study reported that CAD and exposure to noise have a synergistic
increase the risk of subsequent depression. effect on elevating hearing thresholds.[26] Another study

3
Hsu et al. Medicine (2016) 95:44 Medicine

Figure 1. Cummulative incidence comparison of depression for patients with (dashed line) or without (solid line) sensory hearing loss.

suggested that hypertension and diabetes mellitus have a prevalent in the SHL cohort (Table 2). After adjustment for age,
synergistic effect on hearing impairment.[27] Previous studies sex, and comorbidities, hearing loss was observed as an
have reported that patients with hyperlipidemia have a greater independent risk factor for depression (Table 3). The results
risk of noise-induced hearing loss.[28,29] Moderate CKD was are consistent with the results of a US study.[11] Li et al[11]
associated independently with hearing loss.[30] Smoking and reported a significant association between moderate hearing loss
heavy alcohol consumption were related to high-frequency and depression in women, particularly those aged <70 years, but
hearing loss in the Korean population.[31] In our study, not in men or patients aged ≥70 years. These findings are
comorbidities of hypertension, hyperlipidemia, diabetes mellitus, different from our results. Our study revealed that the age-specific
CKD, CAD, and alcohol-related illness were more prevalent in relative risk of depression for the SHL cohort compared with the
the SHL cohort (Table 1). This observation is consistent with the non-SHL cohort was significant in all age groups, except in the
results of previous studies. age group of 50 to 64 years, which also exhibited a trend of an
Depression is associated with several comorbidities, such as increased risk. The risk in patients in different age groups and
CKD, CAD, alcohol-related illness, and generalized anxiety with psychosocial ailments differs, probably because of differ-
disorder, and the association has been clearly illustrated in ences in race, ethnicity, lifestyle, responsibility, and circum-
previous studies.[32–34] Our study revealed that the risk of stances. We found that the relative risk of depression was
depression was higher in patients with comorbid CAD, alcohol- significantly higher in both men and women (Table 3). However,
related illness, and anxiety (Table 2). Furthermore, the rate of while Gopinath et al[36] reported that depression is more common
depression increased with age and was higher in women than in in women than in men with hearing loss, Harada et al[37] reported
man. To minimize the influence from smoking, we have used an that hearing loss is associated with depression in men.
alternative way and adjusted for smoking-related diseases Furthermore, several comorbidities, such as CAD, alcohol-
(including COPD, CAD, stroke, asthma) in our analysis in related illness, and anxiety, were associated with both SHL and
accordance with previous studies.[35] Indeed, we have listed the depression in the present study (Table 3).
association between smoking-related diseases and SHL, and we Our study has some limitations. First, we did not classify the
also noted the COPD, CAD, stroke, and asthma were more severity of hearing impairment and demonstrate its effect on

4
Hsu et al. Medicine (2016) 95:44 www.md-journal.com

Table 2
The incidence and hazard ratio for depression and depression-associated risk factor.
Variable Event PY Rate# Crude HR (95% CI) Adjusted HR† (95% CI)
Sensory hearing loss
No 534 111,744 4.78 1.00 1.00
Yes 273 28,740 9.50 1.99 (1.72, 2.31)‡ 1.73 (1.49, 2.00)‡
Age, y
49 165 38,267 4.31 1.00 1.00
50–64 198 37,654 5.26 1.21 (0.99, 1.49) 0.94 (0.76, 1.17)
65+ 444 64,564 6.88 1.56 (1.30, 1.87)‡ 1.05 (0.85, 1.30)
Sex
Women 370 54,959 6.73 1.33 (1.15, 1.52)‡ 1.35 (1.17, 1.56)‡
Men 437 85,526 5.11 1.00 1.00
Comorbidity
Cirrhosis
No 593 114,776 5.17 1.00 1.00
Yes 214 25,709 8.32 1.60 (1.37, 1.87)‡ 1.18 (1.00, 1.39)
Rheumatoid arthritis
No 804 140,221 5.73 1.00 1.00
Yes 3 264 11.4 1.90 (0.61, 5.92) 1.56 (0.50, 4.90)
Hypertension
No 361 82,152 4.39 1.00 1.00
Yes 446 58,332 7.65 1.71 (1.49, 1.97)‡ 1.09 (0.91, 1.31)
Hyperlipidemia
No 554 110,576 5.01 1.00 1.00
Yes 253 29,908 8.46 1.66 (1.43, 1.93)‡ 1.12 (0.95, 1.33)
Diabetes mellitus
No 684 125,003 5.47 1.00 1.00
Yes 123 15,482 7.94 1.42 (1.17, 1.72)‡ 1.03 (0.84,1 .27)
Asthma
No 709 129,414 5.48 1.00 1.00
Yes 98 11,071 8.85 1.58 (1.28, 1.95)‡ 1.05 (0.83, 1.33)
Chronic kidney disease
No 784 138,005 5.68 1.00 1.00
∗
Yes 23 2480 9.27 1.58 (1.04, 2.39) 1.10 (0.72, 1.68)
Coronary artery disease
No 509 110,882 4.59 1.00 1.00
Yes 298 29,602 10.1 2.16 (1.87, 2.50)‡ 1.52 (1.28, 1.81)‡
Alcohol-related illness
No 774 137,382 5.63 1.00 1.00
Yes 33 3103 10.6 1.81 (1.28, 2.57)‡ 1.61 (1.12, 2.29)†
Anxiety
No 674 133,043 5.07 1.00 1.00
Yes 133 7442 17.9 3.42 (2.84, 4.13)‡ 2.38 (1.95, 2.89)‡
COPD
No 614 118,024 5.20 1.00 1.00
Yes 193 22,460 8.59 1.62 (1.38, 1.91)‡ 1.15 (0.95, 1.38)
Stroke
No 741 133,914 5.53 1.00 1.00
∗
Yes 66 6571 10.0 1.58 (1.28, 1.95)‡ 1.31 (1.01, 1.70)
Medication
Steroid
No 698 127,001 5.50 1.00 1.00
Yes 109 13,483 8.08 1.43 (1.17, 1.75)‡ 1.03 (0.83, 1.27)
Rate#, incidence rate, per 1,000 person-years; Crude HR, relative hazard ratio, Adjusted HR†, multivariable analysis including age, sex, and comorbidities of cirrhosis, hypertension, hyperlipidemia, diabetes
mellitus, asthma, chronic kidney disease, coronary artery disease, alcohol-related illness, anxiety, COPD and stroke, and medication of steroid.
∗
P <0.05.
†
P <0.01.
‡
P <0.001.

depression by group. However, after adjustment for age, sex, and hearing impairment during its progression. This study focused on
comorbidities, the risk of depression was higher in the SHL the effect of objective diagnosis of SHL instead of subjective
cohort than in the non-SHL group, suggesting that SHL may be a symptoms of tinnitus. Third, the diagnosis of depression in our
risk factor for subsequent depression regardless of severity. study was based on ICD-9-CM code. We could not make sure
Second, various studies have established a correlation between whether those patients receive standard diagnostic interview to
tinnitus and psychological illness. Tinnitus typically accompanies make this diagnosis. Even though this limitation could raise the

5
Hsu et al. Medicine (2016) 95:44 Medicine

Table 3
Incidence of depression by age, sex and comorbidity and Cox model measured hazards ratio for patients with sensory hearing loss
compared those without sensory hearing loss.
Sensory hearing loss
No Yes
Variables Event PY Rate# Event PY Rate# Crude HR (95% CI) Adjusted HR† (95% CI)
Age, y
49 101 30,731 3.29 64 7536 8.49 2.58 (1.89, 3.53)‡ 2.22 (1.62, 3.06)‡
50–64 142 30,125 4.71 56 7529 7.44 1.58 (1.16, 2.15)† 1.21 (0.88, 1.67)
65+ 291 50,889 5.72 153 13,676 11.2 1.97 (1.62, 2.39)‡ 1.78 (1.46, 2.17)‡
Sex
Women 256 43,887 5.83 114 11,073 10.3 1.77 (1.42, 2.20)‡ 1.55 (1.24, 1.95)‡
Men 278 67,858 4.10 159 17,668 9.00 2.21 (1.82, 2.68)‡ 1.88 (1.54, 2.29)‡
Comorbidity‡
No 136 47,144 2.88 46 8786 5.24 1.82 (1.30, 2.54)‡ 1.83 (1.31, 2.55)‡
Yes 398 64,600 6.16 227 19,955 11.4 1.86 (1.58, 2.19)‡ 1.87 (1.58, 2.20)‡
Medication
Steroid
No 463 101,656 4.55 235 25,345 9.27 2.05 (1.75, 2.39)‡ 1.78 (1.51, 2.08)‡
∗
Yes 71 10,088 7.04 38 3395 11.2 1.58 (1.07, 2.35) 1.40 (0.94, 2.09)
Rate#, incidence rate, per 1000 person-years; Crude HR, relative hazard ratio; Adjusted HR†, multivariable analysis including age, sex, and comorbidities of cirrhosis, hypertension, hyperlipidemia, diabetes
mellitus, asthma, chronic kidney disease, coronary artery disease, alcohol-related illness, anxiety, COPD and stroke, and medication of steroid; Comorbidity‡, patients with any one of the comorbidities cirrhosis,
hypertension, hyperlipidemia, diabetes mellitus, asthma, chronic kidney disease, coronary artery disease, alcohol-related illness, anxiety, COPD and stroke were classified as the comorbidity group.
∗
P <0.05.
†
P <0.01.
‡
P <0.001.

concern about the probability of false positive or false negative [7] Dawes P, Emsley R, Cruickshanks KJ, et al. Hearing loss and cognition:
the role of hearing AIDS, social isolation and depression. PLoS One
case, our study demonstrated long-term observation to observe
2015;10:e0119616.
the course between hearing loss and depression. Fourth, suicide [8] Abrams TE, Barnett MJ, Hoth A, et al. The relationship between hearing
was prevalent in patients with depression.[38] Even though impairment and depression in older veterans. J Am Geriatr Soc 2006;
suicidal event in depression was not further discussed in our 54:1475–7.
study, we considered this relationship could raise more concern in [9] Chi I, Yip PS, Chiu HF, et al. Prevalence of depression and its correlates in
Hong Kong’s Chinese older adults. Am J Geriatr Psychiatry 2005;
clinical practice, based on our study. Finally, although SHL and 13:409–16.
depression are very frequent, our well-designed study protocol [10] Gomaa MA, Elmagd MH, Elbadry MM, et al. Anxiety and Stress Scale in
and large population cohort have demonstrated the association patients with tinnitus and hearing loss. Eur Arch Otorhinolaryngol
between SHL and depression. After including the analysis of 2014;271:2177–84.
[11] Li CM, Zhang X, Hoffman HJ, et al. Hearing impairment associated
steroid treatment for SHL (Table 3), we further demonstrated the
with depression in US adults, National Health and Nutrition Examina-
possible causal association; however, future prospective studies tion Survey 2005-2010. JAMA Otolaryngol Head Neck Surg 2014;
are required. 140:293–302.
[12] Database NHIR. Taiwan, Available at: http://nhird.nhri.org.tw/en/
index.html (cited in 2015). November 2015.
5. Conclusion [13] Hsu CW, Lin CS, Chen SJ, et al. Risk of type 2 diabetes mellitus in
patients with acute critical illness: a population-based cohort study.
Our data demonstrated that SHL may increase the risk of Intensive Care Med 2016;42:38–45.
subsequent depression. However, information on networks and [14] Wang CC, Chang CT, Lin CL, et al. Hepatitis C virus infection associated
etiologic mechanisms between auditory damage and depression is with an increased risk of deep vein thrombosis: a population-based
scant. Additional prospective clinical and biomedical studies cohort study. Medicine (Baltimore) 2015;94:e1585.
[15] Gao Y, Liu D. Combined intratympanic and systemic use of steroids for
on the relationship between hearing loss and depression are
idiopathic sudden sensorineural hearing loss: a meta-analysis. Eur Arch
warranted to determine the etiopathology. Otorhinolaryngol 2016;273:3699–711.
[16] Strawbridge WJ, Wallhagen MI, Shema SJ, et al. Negative consequences
References of hearing impairment in old age: a longitudinal analysis. Gerontologist
2000;40:320–6.
[1] World Health OrganizationThe Global Burden of Disease: 2004 Update. [17] Kramer SE, Kapteyn TS, Kuik DJ, et al. The association of hearing
2008;Geneva, Switzerland:World Health Organization, 2008. impairment and chronic diseases with psychosocial health status in older
[2] Organization WH. Deafness and hearing loss. March 2015; Available at: age. J Aging Health 2002;14:122–37.
http://www.who.int/mediacentre/factsheets/fs300/en/. November 2015. [18] Mohr PE, Feldman JJ, Dunbar JL. The societal costs of severe to
[3] Organization WH. Depression. April 2016; Available at: http://www. profound hearing loss in the United States. Policy Anal Brief H Ser
who.int/mediacentre/factsheets/fs369/en/. November 2015. 2000;2:1–4.
[4] Mener DJ, Betz J, Genther DJ, et al. Hearing loss and depression in older [19] Dalton DS, Cruickshanks KJ, Klein BE, et al. The impact of hearing loss
adults. J Am Geriatr Soc 2013;61:1627–9. on quality of life in older adults. Gerontologist 2003;43:661–8.
[5] Monzani D, Galeazzi GM, Genovese E, et al. Psychological profile and [20] Haenisch B, Bonisch H. Depression and antidepressants: insights from
social behaviour of working adults with mild or moderate hearing loss. knockout of dopamine, serotonin or noradrenaline re-uptake trans-
Acta Otorhinolaryngol Ital 2008;28:61–6. porters. Pharmacol Ther 2011;129:352–68.
[6] Huang CQ, Dong BR, Lu ZC, et al. Chronic diseases and risk for [21] Helton SG, Lohoff FW. Serotonin pathway polymorphisms and the
depression in old age: a meta-analysis of published literature. Ageing Res treatment of major depressive disorder and anxiety disorders. Pharma-
Rev 2010;9:131–41. cogenomics 2015;16:541–53.

6
Hsu et al. Medicine (2016) 95:44 www.md-journal.com

[22] Ozturk M, Ucar S, Sari F, et al. Possible protective effect of sertraline [30] Vilayur E, Gopinath B, Harris DC, et al. The association between
against cisplatin-induced ototoxicity: an experimental study. Scientific- reduced GFR and hearing loss: a cross-sectional population-based study.
WorldJournal 2013;2013:523480. Am J Kidney Dis 2010;56:661–9.
[23] Rao D, Basura GJ, Roche J, et al. Hearing loss alters serotonergic [31] Lee JS, Choi HG, Jang JH, et al. Analysis of predisposing factors for
modulation of intrinsic excitability in auditory cortex. J Neurophysiol hearing loss in adults. J Korean Med Sci 2015;30:1175–82.
2010;104:2693–703. [32] Palmer S, Vecchio M, Craig JC, et al. Prevalence of depression in chronic
[24] Smith AR, Kwon JH, Navarro M, et al. Acoustic trauma triggers kidney disease: systematic review and meta-analysis of observational
upregulation of serotonin receptor genes. Hear Res 2014;315: studies. Kidney Int 2013;84:179–91.
40–8. [33] Boden JM, Fergusson DM. Alcohol and depression. Addiction 2011;
[25] Lin FR, Thorpe R, Gordon-Salant S, et al. Hearing loss prevalence and 106:906–14.
risk factors among older adults in the United States. J Gerontol A Biol Sci [34] Simon NM. Generalized anxiety disorder and psychiatric comorbidities
Med Sci 2011;66:582–90. such as depression, bipolar disorder, and substance abuse. J Clin
[26] Agrawal Y, Platz EA, Niparko JK. Risk factors for hearing loss in US Psychiatry 2009;70(suppl 2):10–4.
adults: data from the National Health and Nutrition Examination [35] Chang KH, Chang MY, Muo CH, et al. Increased risk of dementia in
Survey, 1999 to 2002. Otol Neurotol 2009;30:139–45. patients exposed to nitrogen dioxide and carbon monoxide: a population-
[27] Rolim LP, Rabelo CM, Lobo IF, et al. Interaction between diabetes based retrospective cohort study. PLoS One 2014;9:e103078.
mellitus and hypertension on hearing of elderly. Codas 2015;27: [36] Gopinath B, Wang JJ, Schneider J, et al. Depressive symptoms in older
428–32. adults with hearing impairments: the Blue Mountains Study. J Am
[28] Chang NC, Yu ML, Ho KY, et al. Hyperlipidemia in noise-induced Geriatr Soc 2009;57:1306–8.
hearing loss. Otolaryngol Head Neck Surg 2007;137:603–6. [37] Harada S, Nishiwaki Y, Michikawa T, et al. Gender difference in the
[29] Sutbas A, Yetiser S, Satar B, et al. Low-cholesterol diet and antilipid relationships between vision and hearing impairments and negative well-
therapy in managing tinnitus and hearing loss in patients with noise- being. Prev Med 2008;47:433–7.
induced hearing loss and hyperlipidemia. Int Tinnitus J 2007;13: [38] Kuo DC, Tran M, Shah AA, et al. Depression and the suicidal patient.
143–9. Emerg Med Clin North Am 2015;33:765–78.