Hearing Research 384 (2019) 107827

Contents lists available at ScienceDirect

Hearing Research
journal homepage: www.elsevier.com/locate/heares

Research Paper

Association between shift work and hearing loss: The Dongfeng-Tongji

cohort study
Dongming Wang a, b, Yun Zhou a, b, Jixuan Ma a, b, Lili Xiao a, b, Limin Cao a, b, Min Zhou a, b,
Weijia Kong c, Zhichao Wang c, Wenzhen Li d, Meian He a, b, Xiaomin Zhang a, b,
Huan Guo a, b, Jing Yuan a, b, Weihong Chen a, b, *
a
Department of Occupational & Environmental Health, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology,
Wuhan, Hubei, 430030, China
b
Key Laboratory of Environment and Health, Ministry of Education & Ministry of Environmental Protection, State Key Laboratory of Environmental Health
(Incubating), School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, 430030, China
c
Department of Otorhinolaryngology, Union Hospital, Tongji Medical College, Huazhong University of Scienceand Technology, Wuhan, Hubei, 430030,
China
d
Department of Social Medicine and Health Management, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology,
Wuhan, Hubei, 430030, China

a r t i c l e i n f o a b s t r a c t

Article history: The association between shift work and hearing loss is unclear. We aimed to evaluate this association in a
Received 12 April 2019 Chinese population independently and in combination with occupational noise. A total of 11,196 par-
Received in revised form ticipants of the Dongfeng-Tongji cohort study were included. Shift work was self-reported and hearing
16 September 2019
loss was defined as a pure-tone mean of 25 dB or higher at 0.5, 1, 2, and 4 kHz in any ear. The Robust
Accepted 23 October 2019
Available online 26 October 2019
Poisson method were used to assess the relationship between shift work and hearing loss, and the
prevalence ratios (PRs) were calculated. Compared to individuals who reported no shift work, the PRs of
bilateral hearing loss were significantly higher for a shift work duration of fewer than 10 years in women,
Keywords:
Shift work
(PR ¼ 1.024, 95% confidence interval [CI] ¼ 1.002e1.053), but not in men (PR ¼ 1.016, 95% confidence
Hearing loss interval [CI] ¼ 0.998e1.035). The association between short duration of shift work and bilateral hearing
Occupational noise loss was just statistically significant in women when those with occupational noise exposure were
excluded (PR ¼ 1.067, 95%CI ¼ 1.015e1.122). When shift work and occupational noise exposure were
combined, the PRs for hearing loss were highest among individuals with the longest shift work (10
years) and longest noise exposure (20 years) durations, whether for bilateral (PR ¼ 1.114, 95%CI ¼ 1.068
e1.162) or any ear (PR ¼ 1.065, 95%CI ¼ 1.034e1.096). A short duration of shift work may be a risk factor
for hearing loss in women and could increase the prevalence of hearing loss when combined with
occupational noise.
© 2019 Published by Elsevier B.V.

1. Introduction of major public health concern. Irregular working hours and

Shift work is common and often unavoidable with about 29% of
workers in the United States experiencing this type of work
schedule (Alterman et al., 2013). Furthermore, about 22% of men
and 11% of women work on shifts that including night work in
Europe (Hansen, 2017). The health consequences of shift work are
* Corresponding author. Department of Occupational and Environmental Health,
School of Public Health, Tongji Medical College, Huazhong University of Science and
Technology, No. 13 Hangkong Road, Wuhan, 430030, PR China.
E-mail address: wchen@mails.tjmu.edu.cn (W. Chen).
different types of work schedules are associated with a range of
adverse health outcomes such as sleep disturbance (Linton et al.,
2015), diabetes mellitus (Pan et al., 2011), coronary heart disease
(Vyas et al., 2012), metabolic disorders (Proper et al., 2016), and
other chronic health problems (Kecklund et al., 2016;
Sigurdardottir et al., 2012).
Hearing loss is also a public health issue, it is reported that the
prevalence of unilateral and bilateral speech-frequency hearing
loss was 6.6% and 7.5% in US adults aged 20e69 years in 2011e2012,
respectively (Hoffman et al., 2017). In occupational settings, Noise-
induced hearing loss (NIHL) is one of the leading chronic health

https://doi.org/10.1016/j.heares.2019.107827
0378-5955/© 2019 Published by Elsevier B.V.
2 D. Wang et al. / Hearing Research 384 (2019) 107827
problems experienced by occupational populations and is one of
the most recorded occupational diseases in Europe and other
countries(Samant et al., 2008). Except for occupational noise,
several other factors are reported to be associated with hearing
loss, including age; sex; race; smoking; metabolic syndromes; and
chronic disease including hypertension, cardiovascular diseases,
and diabetes mellitus (Agrawal et al., 2009; Cruickshanks et al.,
1998). Besides, shift work is also reported to be associated with
metabolic syndromes, hypertension, cardiovascular diseases and
diabetes mellitus (Vetter et al., 2018; Wang et al., 2014).
Few studies have investigated the influence of shift work on
hearing loss. Two previous studies mainly compared the effects of
different types of shift work on hearing loss with small samples to
test the hypothesis that rotating shift work could affect the
continuous exposure time of occupational noise (Chou et al., 2009;
Holzmuller et al., 1990). However, these studies did not assess the
independent effect of shift work on hearing loss after adjusting for
potential confounders. In addition, the potential joint effect of shift
work and occupational noise on hearing loss is not well
understood.
Thus, we conducted a study with large sample size to assess the
effect of shift work on hearing loss, independently and in combi-
nation with occupational noise, based on the assumption that shift
work may affect hearing loss.
2. Materials and methods
2.1. Study population
The data used in this study were obtained from the Dongfeng-
Tongji Cohort Study, the details of which have been reported
elsewhere (Wang et al., 2013). All participants were retired em-
ployees of Dongfeng Motor Corporation (DMC) with unique med-
ical insurance card numbers in the DMC’s health-care service
system. The cohort study began in 2008 and follow-up occurred
every five years. Finally, a total of 27,009 retired employees
completed baseline questionnaires and medical examinations were
included. Five years later, 25,978 individuals (96.2%) completed the
follow-up until October 2013. In 2013, 11,513 participants from the
baseline population underwent audiometric examinations.
In the present study, the 11,513 participants who underwent
audiometric examinations in 2013 were included in our study, we
excluded individuals without auditory examination results (n ¼ 57)
and those who provided incomplete baseline information
(n ¼ 100), leaving 11,196 subjects in the present analysis.
2.2. Auditory measures and ascertainment of hearing loss
Each participant first underwent a general physical and otologic
examination. Pure-tone audiometry was then performed by certi-
fied audiologists in a sound-isolated room with a calibrated pure-
tone audiometer (Micro-DSP ZD21). Air conduction thresholds
were determined for each ear at 0.5, 1, 2, 4 and 8 kHz across an
intensity range of 10 to 120 dB. Those who did not response at
maximum value were coded as the maximum value. Hearing loss at
the speech frequencies was defined as a pure-tone mean of 25 dB or
higher at 0.5, 1, 2, and 4 kHz in any ear (Agrawal et al., 2008).
2.3. Ascertainment of shift work
Shift work was assessed using a self-administrated question-
naire. Participants were asked the following question: “Did you
experience shift work during your work history?” Those who
responded affirmatively were further asked about the duration of
shift work. It was defined as any work schedule involving irregular
arrangements that differed from a standard schedule through
0700/0800 to 1700/1800 at least three times per month for one
year. The participants who reported shift work were divided into
two groups by the median duration of shift work; thus, shift work
was divided into three categories: never (N ¼ 6244), 0-10y
(N ¼ 2519), and 10y (N ¼ 2433).
2.4. Ascertainment of noise exposure assessment
The occupational information of each subject was assessed by
the questionnaire, which contained employment and variables
about the work, such as the corporation, job title, and duration of
each job (Wang et al., 2018). The occupational noise levels for each
job title at each workplace were assessed by qualified institutions
according to the companies’ records. Noise exposure levels for
workplaces outside of DMC were determined according to the job
description and in consultation with local occupational hygienists.
Occupational noise exposure was defined as exposure to normal-
ized continuous A-weighted sound pressure level equivalent to an
8-h per day of 80 dB (A) or above (LAeq, 8h  80 dB(A)) for at least
one year. Noise exposure duration was also divided into four
groups: 0, 1-<10y, 10-<20y, 20y.
2.5. Ethical approval
The study was approved by the Ethics and Human Subject
committee of Tongji Medical College, and Dongfeng General Hos-
pital, DMC. The written informed consents were obtained from all
of the participants.
2.6. Covariates
Information on sociodemographic characteristics (sex, race, and
age), smoking status, drinking status, hypertension, ototoxic med-
ications, and chronic disease history were collected using a ques-
tionnaire in face-to-face interviews performed by trained
interviewers. Individuals currently smoking at least one cigarette
per day for more than half a year were defined as current smokers;
otherwise they were regarded as non-smokers, including ever
smokers and never smokers. Those who currently drank at least
one time per week for more than half a year were defined as cur-
rent drinkers; otherwise they were regarded as non-drinkers,
including ever drinkers and never drinkers. Hypertension was
defined as a blood pressure of at least 140/90 mmHg, self-reported
physician diagnosis of hypertension, or self-reported current use of
antihypertensive medication. Use of ototoxic medication was
defined as the self-reported use of loop diuretics, aminoglycosides,
or non-steroidal anti-inflammatory drugs. Chronic disease histories
diagnosed by a physician were reported by the participants,
including diabetes mellitus, coronary heart disease, myocardial
infarction, and stroke.
2.7. Statistical analysis
The sociodemographic characteristics of the participants were
reported as means (SD) for continuous variables and as numbers
(percentages) for categorical variables. The Robust Poisson method
was performed to evaluate the independent and combined asso-
ciations of shift work and occupational noise exposure with hearing
loss, and the prevalence ratios (PRs) were calculated, as the prev-
alence of hearing loss is high in our study (Petersen et al., 2008).
The associations were further evaluated with stratification by sex,
based on previous reports suggesting that sex may be an important
factor for hearing loss (Helzner et al., 2005). Models were con-
structed using those who were non shift-workers and without
 D. Wang et al. / Hearing Research 384 (2019) 107827 3

occupational noise exposure as the reference group. We chose
covariates according to evidence from published literatures
(Agrawal et al., 2009; Cruickshanks et al., 1998). The covariates
included age, sex, race, smoking status, drinking status, hyperten-
sion, ototoxicity medicine, and chronic disease history (diabetes
mellitus, coronary heart disease, myocardial infarction, and stroke),
and all the covariates were adjusted in the statistical analyses. All
statistical analyses were performed using SAS version 9.2 software
(SAS institute Inc., Cary, NC). The statistical tests were two sided,
and significance was set at P < 0.05.
2.8. Sensitivity analysis
To assess whether excluding participants with occupational
noise exposure could bias the results, a sensitivity analysis of the
effect of shift work on hearing loss was performed to test the sta-
bility and robustness of the association. The prevalence of occu-
pational noise exposure was 34.7% (3891/11,196) among the
participants.
3. Results
The characteristics of the participants included in the analysis
were reported by their categories of hearing loss (Table 1). Among
them, 77.6% (8688/11,196) had speech-frequency hearing loss,
44.2% (4592/11,196) were shift workers, and 34.6% (3878/11,196)
were exposed to occupational noise. Compared to those with
normal hearing, participants with speech-frequency hearing loss
were older and they were more likely to be males; smokers;
drinkers; exposed to occupational noise; and have hypertension,
diabetes mellitus, coronary heart disease, myocardial infarction, or
stroke. These features were more evident in those with bilateral
speech-frequency hearing loss.
3.1. Association between shift work and hearing loss
The effect of shift work on hearing loss was shown in Table S1.
Compared to that among participants without shift work, the
prevalence ratios (PRs) were higher among those with shift work,
including bilateral (PR ¼ 1.034, 95% confidence interval
[CI] ¼ 1.006e1.063) and any-ear (PR ¼ 1.023, 95%CI ¼ 1.003e1.043)
speech-frequency hearing loss. In stratified analysis of sex, the PR
for shift work and bilateral hearing loss was statistically significant
in women (PR ¼ 1.046, 95%CI ¼ 1.003e1.096), but not in men
(PR ¼ 1.021, 95%CI ¼ 0.988e1.056).
When shift work was categorized according to the duration
(Table 2), the PRs of bilateral hearing loss was significantly higher
for those with shift work (PR ¼ 1.021, 95%CI ¼ 1.005e1.038 for shift

Table 1
The characteristics of the participants by hearing loss categories.

Variables Participants (N ¼ 11196) Normal Hearing (N ¼ 2508) Speech-frequency hearing Loss (N ¼ 8688) Unilateral (N ¼ 1800) Bilateral (N ¼ 6888)

Age 67.1 ± 7.4 62.3 ± 6.6 68.5 ± 7.0 65.2 ± 6.4 69.3 ± 6.9
Sex
Male 5060(45.2) 668(26.6) 4392(50.6) 701(38.9) 3691(53.6)
Female 6136(54.8) 1840(73.4) 4296(49.4) 1099(61.1) 3197(46.4)
Race
Han 11025(98.5) 2472(98.6) 8553(98.4) 1756(97.6) 6797(98.7)
others 171(1.5) 36(1.4) 135(1.6) 44(2.4) 91(1.3)
Duration of shift work, year
No 6244(55.8) 1384(55.2) 4860(55.9) 993(55.2) 3867(56.1)
0 < x < 10 2519(22.5) 592(23.6) 1927(22.2) 419(23.3) 1508(21.9)
x  10 2433(21.7) 532(21.2) 1901(21.9) 388(21.5) 1513(22.0)
Smoking
Yes 1569(14.0) 239(9.5) 1330(15.3) 225(12.5) 1105(16.0)
No 9627(86.0) 2269(90.5) 7358(84.7) 1575(87.5) 5783(84.0)
Drinking
Yes 2616(23.4) 489(19.5) 2127(24.5) 413(22.9) 1714(24.9)
No 8580(76.6) 2019(80.5) 6561(75.5) 1387(77.1) 5174(75.1)
Hypertension
Yes 7739(69.1) 1518(60.5) 6221(71.6) 1189(66.1) 5032(73.0)
No 3457(30.9) 990(39.5) 2467(28.4) 611(33.9) 1856(27.0)
Diabetes mellitus
Yes 1759(15.7) 342(13.6) 1417(16.3) 272(15.1) 1145(16.6)
No 9437(84.3) 2166(86.4) 7271(83.7) 1528(84.9) 5743(83.4)
Coronary heart disease
Yes 2030(18.1) 335(13.4) 1695(19.5) 276(15.3) 1419(20.6)
No 9166(81.9) 2173(86.6) 6993(80.5) 1524(84.7) 5469(79.4)
Myocardial infarction
Yes 396(3.5) 46(1.8) 350(4.0) 44(2.4) 306(4.4)
No 10800(96.5) 2462(98.2) 8338(96.0) 1756(97.6) 6582(95.6)
Stroke
Yes 529(4.7) 80(3.2) 449(5.2) 58(3.2) 391(5.7)
No 10667(95.3) 2428(96.8) 8239(94.8) 1742(96.8) 6497(94.3)
Ototoxic medicine
Yes 2770(24.7) 615(24.5) 2155(24.8) 464(25.8) 1691(24.5)
No 8426(75.3) 1893(75.5) 6533(75.2) 1336(74.2) 5197(75.5)
Occupational noise exposure
Yes 3878(34.6) 841(33.5) 3037(35.0) 624(34.7) 2413(35.0)
No 7318(65.4) 1667(66.5) 5651(65.0) 1176(65.3) 4475(65.0)
Exposure duration of occupational noise
No 7305(65.3) 1666(66.4) 5639(64.9) 1176(65.3) 4463(64.8)
1  year<10 944(8.4) 253(10.1) 691(8.0) 184(10.2) 507(7.4)
10  year<20 1150(10.3) 271(10.8) 879(10.1) 194(10.8) 685(9.9)
year  20 1797(16.0) 318(12.7) 1479(17.0) 246(13.7) 1233(17.9)
4 D. Wang et al. / Hearing Research 384 (2019) 107827
Table 2
Prevalence ratios (95% CIs) of hearing loss by duration of shift work in total population (N ¼ 11196).
Shift work Unilateral(N ¼ 1800)a
Total
No ref
1  year<10 0.984(0.937e1.033)
year  10 0.968(0.878e1.067)
Male
No ref
1  year<10 0.981(0.909e1.058)
year  10 0.962(0.827e1.120)
Female
No ref
1  year<10 0.981(0.920e1.046)
year  10 0.962(0.847e1.094)
Adjusted for age, sex, race, occupational noise exposure, smoking status, drinking status, hypertension, ototoxicity medicine, chronic diseases (diabetes mellitus,
coronary heart disease, myocardial infarction and stroke).
a
Pure-tone mean of 25 dB or higher at 0.5, 1, 2, and 4 kHz in one ear only.
b
Pure-tone mean of 25 dB or higher at 0.5, 1, 2, and 4 kHz in both ears.
c
Pure-tone mean of 25 dB or higher at 0.5, 1, 2, and 4 kHz in any ear, including one ear only and both ears.
work duration of 1e10 years; PR ¼ 1.043, 95%CI ¼ 1.011e1.077 for
shift work duration of more than 10 years), the results were similar
for any ear hearing loss (PR ¼ 1.013, 95%CI ¼ 1.002e1.024 for shift
work duration of 1e10 years; PR ¼ 1.026, 95%CI ¼ 1.003e1.048 for
shift work duration of more than 10 years). In stratified analysis of
sex, it was just statistically significant for a shift work duration of
fewer than 10 in women (PR ¼ 1.024, 95%CI ¼ 1.002e1.053).
The results of a sensitivity analysis were nearly the same after
excluding participants with occupational noise exposure (Table S2).
A shift work duration of fewer than 10 years was associated with
bilateral (PR ¼ 1.057, 95%CI ¼ 1.015e1.101) and any-ear (PR ¼ 1.036,
95%CI ¼ 1.007e1.066) hearing loss, and the association for bilateral
hearing loss was just statistically significant in women (PR ¼ 1.067,
95%CI ¼ 1.015e1.122).
3.2. Combined effect of shift work and occupational noise exposure
on hearing loss
We further explored the combined association of occupational
noise exposure and shift work on any-ear and both-ear hearing loss
(Fig. S1). Despite bilateral or any-ear hearing loss, the PRs were
highest among those exposed to occupational noise and shift work,
compared to participants without shift work and no occupational
noise exposure.
The effects of shift work duration and noise exposure duration
on bilateral and any-ear hearing loss were further assessed to test
the dose-response relationship (Table 3 and Table 4). The PRs were
highest among individuals with the longest noise exposure (20
years) and shift work durations (10 years) (PR ¼ 1.114, 95%
CI ¼ 1.068e1.162, bilateral; PR ¼ 1.065, 95%CI ¼ 1.034e1.096, any
ear). Sex-specific analyses revealed that the PR for bilateral hearing
loss was higher in men (PR ¼ 1.117, 95%CI ¼ 1.064e1.173) than that
in women (PR ¼ 1.101, 95%CI ¼ 1.023e1.185).
3.3. Association between occupational noise and hearing loss
To confirm the results of our study, the associations between
occupational noise exposure and hearing loss were also explored
(Table S3 and Table S4). Compared to those participants that were
not exposed to occupational noise, the PR of bilateral hearing loss
increased with the increase of noise exposure duration and got the
highest in the longest noise exposure duration group (PR ¼ 1.104,
95%CI ¼ 1.067e1.141). Stratified analyses revealed that the PR be-
tween the longest noise exposure duration (20 years) and bilat-
eral hearing loss was higher in men (PR ¼ 1.129, 95%
Bilateral(N ¼ 6888)b Any ear(N ¼ 8688)c
ref ref
1.021(1.005e1.038) 1.013(1.002e1.024)
1.043(1.011e1.077) 1.026(1.003e1.048)
ref ref
1.016(0.998e1.035) 1.010(0.998e1.022)
1.033(0.995e1.071) 1.019(0.995e1.045)
ref ref
1.024(1.002e1.053) 1.013(0.994e1.032)
1.049(0.993e1.108) 1.026(0.988e1.066)
CI ¼ 1.089e1.171). The results for any ear hearing loss were similar
as those for bilateral hearing loss.
4. Discussion
This study identified a potential association between shift work
and hearing loss, especially bilateral hearing loss. This finding was
sex-specific, it was just statistically significant in women and only
among women with a short duration of shift work. This result was
consistent with those obtained after excluding individuals with
occupational noise exposure. To confirm the results, we also
investigated the association between occupational noise exposure
and hearing loss, and verified the positive relationship among both
men and women, which could explain the reliability of our study.
The underlying mechanism was not clear. Shift work is associ-
ated with diabetes mellitus (Vetter et al., 2018), metabolic syn-
drome (Wang et al., 2014), obesity (Sun et al., 2018), hypertension
(Manohar et al., 2017) and cardiovascular diseases (Torquati et al.,
2017) through the endocrine pathway, as we have previously re-
ported (Guo et al., 2013, 2015). Recent published studies reported
that diabetes mellitus (Kim et al., 2017), metabolic syndrome
(Aghazadeh-Attari et al., 2017), obesity (Curhan et al., 2013), hy-
pertension (Lin et al., 2016) and cardiovascular diseases (Tan et al.,
2018) could increase the risk of hearing loss. The microvascular and
neuropathic complications in multiple organ systems that affect
patients with diabetes mellitus may affect the inner ear (Friedman
et al., 1975; Vinik et al., 2003), including microangiopathic changes
in the capillaries of the stria vascularis, such as narrowing of the
internal auditory artery, thickening of the basilar membrane,
demyelination of the auditory nerve and loss of spiral ganglion
neurons and organ of Corti cells (Akinpelu et al., 2014; Fukushima
et al., 2006). In addition, hypertension and cardiovascular dis-
eases may compromise the vascular supply to the stria vascularis
(Friedland et al., 2009). Moreover, shift work-induced obesity-
related oxidative stress and resultant damage to the auditory
epithelia may also contribute to hearing loss (Loffredo et al., 2012).
Further studies are warranted to investigate the potential biological
mechanism and difference between sexes.
The sex-specific relationship between shift work and hearing
loss was an interesting finding, the reasons for which remain un-
clear. The result implied that the association was statistically sig-
nificant for females, but not for males. This indicates that female
shift workers should pay more attention to the prevention of
hearing loss, and provides a clue for future study of how the bio-
logical mechanisms of shift work and hearing loss are affected by
 D. Wang et al. / Hearing Research 384 (2019) 107827 5

Table 3
Prevalence ratios (95% CIs) of hearing loss by combined categories of occupational noise exposure and duration of shift work (bilateral).

Shift work Noise

No 1  year<10 10  year<20 year20

Total
No ref 1.018(1.011e1.025) 1.027(1.017e1.038) 1.037(1.022e1.051)
1  year<10 1.057(1.015e1.101) 1.055(1.033e1.078) 1.065(1.039e1.092) 1.075(1.045e1.105)
year  10 1.028(0.997e1.049) 1.094(1.056e1.133) 1.104(1.062e1.148) 1.114(1.068e1.162)
Male
No ref 1.019(1.010e1.027) 1.028(1.016e1.041) 1.038(1.021e1.055)
1  year<10 1.015(0.982e1.049) 1.057(1.031e1.083) 1.067(1.037e1.098) 1.076(1.042e1.122)
year  10 1.030(0.963e1.101) 1.096(1.053e1.142) 1.107(1.058e1.157) 1.117(1.064e1.173)
Female
No ref 1.016(1.004e1.029) 1.024(1.006e1.043) 1.033(1.008e1.058)
1  year<10 1.067(1.015e1.122) 1.049(1.012e1.089) 1.058(1.014e1.104) 1.066(1.015e1.120)
year  10 1.033(0.998e1.059) 1.084(1.019e1.152) 1.092(1.021e1.169) 1.101(1.023e1.185)

Adjusted for age, sex, race, smoking status, drinking status, hypertension, ototoxicity medicine, chronic diseases (diabetes mellitus, coronary heart disease, myocardial
infarction and stroke).
Hearing loss was defined as pure-tone mean of 25 dB or higher at 0.5, 1, 2, and 4 kHz in both ears.

Table 4
Prevalence ratios (95% CIs) of hearing loss by combined categories of occupational noise exposure and duration of shift work (any ear).

Shift work Noise

No 1  year<10 10  year<20 year20

Total
No ref 1.011(1.006e1.016) 1.016(1.008e1.023) 1.021(1.011e1.031)
1  year<10 1.036(1.007e1.066) 1.032(1.017e1.047) 1.037(1.020e1.055) 1.043(1.023e1.063)
year  10 1.018(0.993e1.032) 1.054(1.028e1.080) 1.059(1.031e1.088) 1.065(1.034e1.096)
Male
No ref 1.009(1.004e1.015) 1.014(1.006e1.022) 1.019(1.008e1.030)
1  year<10 1.016(0.999e1.033) 1.028(1.012e1.045) 1.033(1.014e1.053) 1.038(1.016e1.061)
year  10 1.033(0.999e1.067) 1.048(1.020e1.076) 1.053(1.022e1.084) 1.058(1.024e1.092)
Female
No ref 1.011(1.003e1.020) 1.017(1.004e1.030) 1.022(1.005e1.040)
1  year<10 1.028(0.981e1.077) 1.034(1.008e1.060) 1.039(1.009e1.071) 1.045(1.010e1.081)
year  10 1.014(0.991e1.038) 1.057(1.013e1.102) 1.062(1.014e1.113) 1.068(1.015e1.124)

Adjusted for age, sex, race, smoking status, drinking status, hypertension, ototoxicity medicine, chronic diseases (diabetes mellitus, coronary heart disease, myocardial
infarction and stroke).
Hearing loss was defined as pure-tone mean of 25 dB or higher at 0.5, 1, 2, and 4 kHz in any ear, including one ear only and both ears.

sex. These biological mechanisms are complex, and comprehensive
research is needed. The possible adverse effect of circadian
disruption owing to shift work may influence the regulation of
estrogens through the hypothalamic-pituitary-gonadal axis
(Dumont et al., 2014; Gomez-Acebo et al., 2015; Papantoniou et al.,
2015), which could contribute to the sex difference between shift
work and hearing loss. Furthermore, the increased risk in women
between shift work and obesity (Sun et al., 2018) and sex differ-
ences in insulin resistance could also contribute to this result (Qiu
et al., 2018). Longer shift work was not significantly associated with
hearing loss, a finding that may be related to the continuous
adaption of the circadian system (Sun et al., 2018), resulting in a
gradually weakened effect of shift work on hearing loss. However,
the potential mechanism requires confirmation in future studies.
As shift work and occupational noise exposure were not asso-
ciated with unilateral hearing loss, we explored their joint effect on
bilateral and any-ear hearing loss. The combination of shift work
and occupational noise resulted in the highest PRs among in-
dividuals with the longest noise exposure (20 years) and shift
work (10 years) durations, regardless of bilateral or any-ear
hearing loss. Meanwhile, the PR was higher in men than that in
women for bilateral hearing loss. The combined sex-specific results
were almost the opposite of the independent association between
shift work and hearing loss mentioned above. Occupational noise
may be the most direct risk factor for hearing loss in occupational
settings and mainly induced bilateral hearing loss. It also indicated
that the independent effect of occupational noise on bilateral
hearing loss was greater than that of shift work (1.055 vs 1.034).
Furthermore, the relationship between occupational noise and
bilateral hearing loss in men was greater than that between shift
work and bilateral hearing loss in women (1.080 vs 1.046), the
special job may be a contributing factor, such as military service for
males. This may explain the inverse sex-specific results of the
combined effect of shift work and occupational noise and the in-
dependent effect of shift work on hearing loss. In addition, the
association between the longest noise exposure duration (20
years) and bilateral hearing loss in men was also greater than that
for a shorter shift work duration in women (1.129 vs 1.024),
resulting in the highest combined PR for the longest noise exposure
(20 years) and longest shift work (10 years) durations. However,
the combined PR of shift work and occupational noise exposure was
still higher than the independent effect of occupational noise,
which indicated that shift work could also increase the prevalence
of hearing loss when combined with occupational noise.
This study has some strengths. It evaluated the association be-
tween shift work and hearing loss in a large sample size, which has
rarely been assessed before. However, several limitations should be
noted. Firstly, it was not a prospective study design; thus, we could
not explore the causal relationship between shift work and hearing
loss. Future follow-up is necessary to assess this association. Sec-
ondly, the shift work information was self-reported. However, the
validity of self-reported exposure to shift work was verified in
6 D. Wang et al. / Hearing Research 384 (2019) 107827
previous studies (Harma et al., 2017). Thirdly, some other con-
founders were still not included in the analyses, such as recrea-
tional noise exposure, hearing protection devices, etc. However,
most of the employers might not wear hearing protection devices
according to the company records, as the protection consciousness
of the workers may be poor in the past.
5. Conclusions
In summary, our data provide evidence that shift work was
independently associated with bilateral hearing loss in this large
sample study of middle-aged and elderly Chinese adults, a finding
that was just statistically significant among women. When com-
bined with occupational noise exposure, shift work could increase
the prevalence of noise-induced hearing loss. Future studies are
needed to evaluate the potential mechanisms of these findings.
Funding
The study was supported by the National Natural Science
Foundation of China (81903291), China Postdoctoral Science
Foundation (2019T120666, 2018M640705, 2019T120665) and the
Fundamental Research Funds for the Central Universities,
HUST2016JCTD116. The funder did not play any role in study
design; in the collection, analysis, and interpretation of data; in the
writing of the report; nor in the preparation, review, or approval of
the manuscript.
Declaration of competing interest
None.
Acknowledgments
The contributions of all the participants, staffs of the Health
Examination Center of the Dongfeng Central Hospital and the
Medical Insurance Center of DMC, and all members of study team
are greatly acknowledged.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.heares.2019.107827.
References
Aghazadeh-Attari, J., Mansorian, B., Mirza-Aghazadeh-Attari, M., Ahmadzadeh, J.,
Mohebbi, I., 2017. Association between metabolic syndrome and sensorineural
hearing loss: a cross-sectional study of 11,114 participants. Diabetes Metab.
Syndr. Obes. 10, 459e465.
Agrawal, Y., Platz, E.A., Niparko, J.K., 2008. Prevalence of hearing loss and differ-
ences by demographic characteristics among US adults: data from the national
health and nutrition examination survey, 1999-2004. Arch. Intern. Med. 168,
1522e1530.
Agrawal, Y., Platz, E.A., Niparko, J.K., 2009. Risk factors for hearing loss in US adults:
data from the national health and nutrition examination survey, 1999 to 2002.
Otol. Neurotol. 30, 139e145.
Akinpelu, O.V., Ibrahim, F., Waissbluth, S., Daniel, S.J., 2014. Histopathologic changes
in the cochlea associated with diabetes mellitus–a review. Otol. Neurotol. 35,
764e774.
Alterman, T., Luckhaupt, S.E., Dahlhamer, J.M., Ward, B.W., Calvert, G.M., 2013.
Prevalence rates of work organization characteristics among workers in the
U.S.: data from the 2010 national health interview survey. Am. J. Ind. Med. 56,
647e659.
Chou, Y.F., Lai, J.S., Kuo, H.W., 2009. Effects of shift work on noise-induced hearing
loss. Noise Health 11, 185e188.
Cruickshanks, K.J., Klein, R., Klein, B.E., Wiley, T.L., Nondahl, D.M., Tweed, T.S., 1998.
Cigarette smoking and hearing loss: the epidemiology of hearing loss study.
J. Am. Med. Assoc. 279, 1715e1719.
Curhan, S.G., Eavey, R., Wang, M., Stampfer, M.J., Curhan, G.C., 2013. Body mass
index, waist circumference, physical activity, and risk of hearing loss in women.
Am. J. Med. 126, 1142 e1-8.
Dumont, M., Paquet, J., 2014. Progressive decrease of melatonin production over
consecutive days of simulated night work. Chronobiol. Int. 31, 1231e1238.
Friedland, D.R., Cederberg, C., Tarima, S., 2009. Audiometric pattern as a predictor of
cardiovascular status: development of a model for assessment of risk. The
Laryngoscope 119, 473e486.
Friedman, S.A., Schulman, R.H., Weiss, S., 1975. Hearing and diabetic neuropathy.
Arch. Intern. Med. 135, 573e576.
Fukushima, H., Cureoglu, S., Schachern, P.A., Paparella, M.M., Harada, T., Oktay, M.F.,
2006. Effects of type 2 diabetes mellitus on cochlear structure in humans. Arch.
Otolaryngol. Head Neck Surg. 132, 934e938.
Gomez-Acebo, I., Dierssen-Sotos, T., Papantoniou, K., Garcia-Unzueta, M.T., Santos-
Benito, M.F., Llorca, J., 2015. Association between exposure to rotating night
shift versus day shift using levels of 6-sulfatoxymelatonin and cortisol and
other sex hormones in women. Chronobiol. Int. 32, 128e135.
Guo, Y., Liu, Y., Huang, X., Rong, Y., He, M., Wang, Y., Yuan, J., Wu, T., Chen, W., 2013.
The effects of shift work on sleeping quality, hypertension and diabetes in
retired workers. PLoS One 8, e71107.
Guo, Y., Rong, Y., Huang, X., Lai, H., Luo, X., Zhang, Z., Liu, Y., He, M., Wu, T., Chen, W.,
2015. Shift work and the relationship with metabolic syndrome in Chinese aged
workers. PLoS One 10, e0120632.
Hansen, J., 2017. Night shift work and risk of breast cancer. Curr. Environ. Health
Rep. 4, 325e339.
Harma, M., Koskinen, A., Ropponen, A., Puttonen, S., Karhula, K., Vahtera, J.,
Kivimaki, M., 2017. Validity of self-reported exposure to shift work. Occup.
Environ. Med. 74, 228e230.
Helzner, E.P., Cauley, J.A., Pratt, S.R., Wisniewski, S.R., Zmuda, J.M., Talbott, E.O., de
Rekeneire, N., Harris, T.B., Rubin, S.M., Simonsick, E.M., Tylavsky, F.A.,
Newman, A.B., 2005. Race and sex differences in age-related hearing loss: the
health, aging and body composition study. J. Am. Geriatr. Soc. 53, 2119e2127.
Hoffman, H.J., Dobie, R.A., Losonczy, K.G., Themann, C.L., Flamme, G.A., 2017.
Declining prevalence of hearing loss in US adults aged 20 to 69 years. JAMA
Otolaryngol. Head Neck Surg. 143, 274e285.
Holzmuller, M., Seibt, A., Jakubowski, A., Friedrichsen, G., 1990. Studies on the
combined effects of shift work and noise on permanent hearing loss. Z. Gesamte
Hyg. 36, 501e502.
Kecklund, G., Axelsson, J., 2016. Health consequences of shift work and insufficient
sleep. BMJ 355, i5210.
Kim, M.B., Zhang, Y., Chang, Y., Ryu, S., Choi, Y., Kwon, M.J., Moon, I.J., Deal, J.A.,
Lin, F.R., Guallar, E., Chung, E.C., Hong, S.H., Ban, J.H., Shin, H., Cho, J., 2017.
Diabetes mellitus and the incidence of hearing loss: a cohort study. Int. J. Epi-
demiol. 46, 717e726.
Lin, B.M., Curhan, S.G., Wang, M., Eavey, R., Stankovic, K.M., Curhan, G.C., 2016.
Hypertension, diuretic use, and risk of hearing loss. Am. J. Med. 129, 416e422.
Linton, S.J., Kecklund, G., Franklin, K.A., Leissner, L.C., Sivertsen, B., Lindberg, E.,
Svensson, A.C., Hansson, S.O., Sundin, O., Hetta, J., Bjorkelund, C., Hall, C., 2015.
The effect of the work environment on future sleep disturbances: a systematic
review. Sleep Med. Rev. 23, 10e19.
Loffredo, L., Martino, F., Carnevale, R., Pignatelli, P., Catasca, E., Perri, L.,
Calabrese, C.M., Palumbo, M.M., Baratta, F., Del Ben, M., Angelico, F., Violi, F.,
2012. Obesity and hypercholesterolemia are associated with NOX2 generated
oxidative stress and arterial dysfunction. J. Pediatr. 161, 1004e1009.
Manohar, S., Thongprayoon, C., Cheungpasitporn, W., Mao, M.A., Herrmann, S.M.,
2017. Associations of rotational shift work and night shift status with hyper-
tension: a systematic review and meta-analysis. J. Hypertens. 35, 1929e1937.
Pan, A., Schernhammer, E.S., Sun, Q., Hu, F.B., 2011. Rotating night shift work and
risk of type 2 diabetes: two prospective cohort studies in women. PLoS Med. 8,
e1001141.
Papantoniou, K., Pozo, O.J., Espinosa, A., Marcos, J., Castano-Vinyals, G., Basagana, X.,
Juanola Pages, E., Mirabent, J., Martin, J., Such Faro, P., Gasco Aparici, A.,
Middleton, B., Skene, D.J., Kogevinas, M., 2015. Increased and mistimed sex
hormone production in night shift workers. Cancer Epidemiol. Biomark. Prev.
24, 854e863.
Petersen, M.R., Deddens, J.A., 2008. A comparison of two methods for estimating
prevalence ratios. BMC Med. Res. Methodol. 8, 9.
Proper, K.I., van de Langenberg, D., Rodenburg, W., Vermeulen, R.C., van der
Beek, A.J., van Steeg, H., van Kerkhof, L.W., 2016. The relationship between shift
work and metabolic risk factors: a systematic review of longitudinal studies.
Am. J. Prev. Med. 50, e147ee157.
Qiu, J., Bosch, M.A., Meza, C., Navarro, U.V., Nestor, C.C., Wagner, E.J.,
Ronnekleiv, O.K., Kelly, M.J., 2018. Estradiol protects proopiomelanocortin
neurons against insulin resistance. Endocrinology 159, 647e664.
Samant, Y., Parker, D., Wergeland, E., Wannag, A., 2008. The Norwegian labour in-
spectorate’s registry for work-related diseases: data from 2006. Int. J. Occup.
Environ. Health 14, 272e279.
Sigurdardottir, L.G., Valdimarsdottir, U.A., Fall, K., Rider, J.R., Lockley, S.W.,
Schernhammer, E., Mucci, L.A., 2012. Circadian disruption, sleep loss, and
prostate cancer risk: a systematic review of epidemiologic studies. Cancer
Epidemiol. Biomark. Prev. 21, 1002e1011.
Sun, M., Feng, W., Wang, F., Li, P., Li, Z., Li, M., Tse, G., Vlaanderen, J., Vermeulen, R.,
Tse, L.A., 2018. Meta-analysis on shift work and risks of specific obesity types.
Obes. Rev. 19, 28e40.
Tan, H.E., Lan, N.S.R., Knuiman, M.W., Divitini, M.L., Swanepoel, D.W., Hunter, M.,
Brennan-Jones, C.G., Hung, J., Eikelboom, R.H., 2018. Associations between
 D. Wang et al. / Hearing Research 384 (2019) 107827
cardiovascular disease and its risk factors with hearing loss-A cross-sectional
analysis. Clin. Otolaryngol. 43, 172e181.
Torquati, L., Mielke, G.I., Brown, W.J., Kolbe-Alexander, T., 2017. Shift work and the
risk of cardiovascular disease. A systematic review and meta-analysis including
dose-response relationship. Scand. J. Work Environ. Health 44, 229e238.
Vetter, C., Dashti, H.S., Lane, J.M., Anderson, S.G., 2018. Night shift work, genetic risk,
and type 2 diabetes in the UK biobank. Diabetes Care 41, 762e769.
Vinik, A.I., Maser, R.E., Mitchell, B.D., Freeman, R., 2003. Diabetic autonomic neu-
ropathy. Diabetes Care 26, 1553e1579.
Vyas, M.V., Garg, A.X., Iansavichus, A.V., Costella, J., Donner, A., Laugsand, L.E.,
Janszky, I., Mrkobrada, M., Parraga, G., Hackam, D.G., 2012. Shift work and
vascular events: systematic review and meta-analysis. BMJ 345, e4800.
7
Wang, D., Zhou, M., Li, W., Kong, W., Wang, Z., Guo, Y., Zhang, X., He, M., Guo, H.,
Chen, W., 2018. Occupational noise exposure and hypertension: the Dongfeng-
Tongji cohort study. J. Am. Soc. Hypertens. 12, 71e79 e5.
Wang, F., Zhang, L., Zhang, Y., Zhang, B., He, Y., Xie, S., Li, M., Miao, X., Chan, E.Y.,
Tang, J.L., Wong, M.C., Li, Z., Yu, I.T., Tse, L.A., 2014. Meta-analysis on night shift
work and risk of metabolic syndrome. Obes. Rev. 15, 709e720.
Wang, F., Zhu, J., Yao, P., Li, X., He, M., Liu, Y., Yuan, J., Chen, W., Zhou, L., Min, X.,
Fang, W., Liang, Y., Wang, Y., Wei, S., Liu, J., Miao, X., Lang, M., Jiang, X., Zhang, P.,
Li, D., Lu, C., Wang, X., Shi, W., Zheng, J., Guo, H., Zhang, X., Yang, H., Hu, F.B.,
Wu, T., 2013. Cohort Profile: the Dongfeng-Tongji cohort study of retired
workers. Int. J. Epidemiol. 42, 731e740.