Associations of Earphone Use with Tinnitus and Anxiety/Depression.

OBJECTIVE: This study aimed to investigate the association of earphone use with audiologic and psychologic factors.
MATERIALS AND METHODS: Korea National Health and Nutrition Examination Survey 2010-2012 data were collected for participants aged ≥12 years old with earphone use ≥1 hour/day. They were matched to a control group for age, sex, income, and education level. The relationship between earphone use and the hearing thresholds of pure-tone audiometry, tinnitus, and psychologic factors such as depression and anxiety, and other quality of life variables was analyzed using multiple logistic regression tests with complex sampling.
RESULTS: Among the participants, 22.9% (449/1955) of earphone users and 18.1% (355/1600) of control participants had tinnitus (P < 0.001). Earphone users showed 1.27-times higher odds for tinnitus (95% confidence interval [CI] = 1.09-1.50, P = 0.003). Moreover, 6.5% (128/1955) of earphone users and 5.0% (97/1600) of control participants had anxiety and depressive symptoms (P = 0.033). Earphone users showed 1.32-times higher odds for anxiety and depressive symptoms (95% CI = 1.14-1.52, P = 0.040). Nevertheless, the hearing thresholds were comparable between earphone users and control participants.
CONCLUSION: Earphone use was associated with tinnitus and anxiety or depressive symptoms.

INTRODUCTION

The use of personal listening devices via earphones is widespread worldwide. In particular, the young population is extensively exposed to earphone use, with as high as 84% of adolescents using earphones.[1] Because the earphone is generally used to concentrate on listening materials, earphone users often select high sound volumes to overcome the noise in the environment.[2] The preferred listening levels were estimated to be approximately 82 (standard deviation [SD] = 9) dBA in calm conditions, whereas they were 89 (SD = 9) dBA in noisy conditions.[2] Therefore, many earphone users experience subjective loudness of sounds and hearing discomfort. A study showed that approximately 49.6% (95% confidence interval [CI] = 44.4–54.4) of earphone users reported loudness of sounds.[1] In addition, many earphone users wear earphones for long durations without intervals. These risky patterns of earphone use could be implicated in hearing problems. Indeed, hearing loss and tinnitus in earphone users were reported in the previous studies.[34]

Tinnitus is one of the most prevalent audiologic discomforts. Its prevalence was estimated to be approximately 11.9% to 30.3% worldwide.[5] In Korea, the prevalence of tinnitus was reported to be approximately 20.7% in the adult population and 17.7% in the 12- to 19-year-old population.[67] Various factors are associated with tinnitus.[89] Its incidence increases with aging and hearing loss in the general population.[10] In the adolescent population, hearing loss and noise exposure were related to higher odds for tinnitus (odds ratio [OR] = 2.39, 95% CI = 1.48–3.87 for hearing loss and OR = 11.35, 95% CI = 1.87–68.77 for noise exposure).[11] Moreover, medical histories of cardiovascular disorders and diabetes were related to tinnitus.[1012] Tinnitus could, in turn, cause psychologic problems and impair the quality of life.[1314] Approximately 40% to 60% and 15% to 60% of adult patients with severe tinnitus were estimated to have depression and anxiety disorders diagnosed with the Structured Clinical Interview for DSM-IV Axis I Disorders and the ICD-9-CM codes, respectively.[1516] A cross-sectional cohort study demonstrated that tinnitus patients had more depressive and anxiety symptoms based on a home interview than control participants (difference = 0.20, 95% CI = 0.11–0.28 for depressive symptom and difference = 0.15, 95% CI = 0.08–0.22 for anxiety).[17] Especially, the patients with tinnitus distress showed as high as 4.8-fold higher odds for development of depression surveyed by questionnaires (95% CI = 3.5–6.7, P < 0.001).[13] Therefore, the impact of earphone use on tinnitus is expected.

A few cross-sectional studies reported hearing problems following chronic earphone use.[18] However, most previous studies included small study populations and lacked suitable controls. In addition, tinnitus is frequently associated with hearing loss. Thus, hearing loss should be adjusted for to prevent its interference on the association between earphone use and tinnitus. A national population cohort study reported 1.19-times higher odds for hearing loss with every 1 hour increase in earphone usage (95% CI = 1.01–1.41).[19] However, they did not consider the degree of hearing loss and binarily categorized the hearing loss as the average pure-tone hearing thresholds at 2, 3, and 4 kHz.

The present study hypothesized that earphone use in noisy conditions might have an adverse impact on tinnitus and the quality of life, including anxiety or depression. To prove this hypothesis, a nationwide, representative cohort population was used. The randomly selected control group was matched to the earphone group for both demographic and socioeconomic factors. To minimize confounders, the medical histories were considered and pure-tone audiometry was evaluated. To exclude other causes of hearing loss, a tympanic endoscopic examination was conducted and only participants with normal findings on tympanic endoscopy were included.

MATERIALS AND METHODS

Ethical considerations

The present study was approved by the Korean Centers for Disease Control and Prevention (IRB No. 2010-02CON-21-C, 2011-02CON-06-C, and 2012-01EXP-01-2C) and the ethics committee of CHA University (2018-06-008-006). Before the survey, all participants or legal representatives of minors were explained about the study, and they provided written informed consent. The data source for the present study was the Korea National Health and Nutrition Examination Survey (KNHANES), which included participants’ representative of the South Korean population. The survey included a health interview and physical examinations.

Among a total of 25,534 participants in the KNHANES from 2010 to 2012, those aged ≥12 years old (maximum age: 85 years old) and with pure-tone audiometry data were included in this study. Participants who did not undergo the earphone use survey (n = 1913), could not remember their earphone use history (n = 44), had no response on the earphone use survey (n = 99), and were younger than 12 years old (n = 3560) were excluded. In addition, participants with abnormal tympanic membrane findings on tympanic endoscopy (n = 2140) and without data on demographic variables or medical history (n = 1614) were excluded. Among the remaining 16,164 participants, those who used earphones in loud conditions for ≥1 hour/day were selected for the study. The participants who used earphones were included in the earphone group and were matched in a 1:1 ratio to the control group of participants for age, sex, income, and education level by using propensity scores. Ultimately, 1955 participants each were included in the earphone group and control group [Figure 1].

Figure 1

Schematic illustration of the participant selection process that was used in the present study. Among a total of 25,534 participants, 1955 earphone users and 1955 control participants were analyzed.

Variables

The potential-associated factors of noise exposure or tinnitus were examined including education level, obesity, medical histories of diabetes mellitus, hypertension, dyslipidemia, and depression.[122021] Because tinnitus was associated with hearing level and advanced age, age and audiometric test were considered as covariates in this study.[12] In addition, education level was reported to be an associated factor for tinnitus, in that adjusted in the present study.[20] Education level was surveyed through questionnaires and categorized into four groups: (1) no more than primary school, (2) lower secondary, (3) higher secondary, and (4) post-secondary. The level of income was collected based on the nationally registered data and also classified into four groups: (1) low, (2) low-intermediate, (3) high-intermediate, and (4) high. The body mass index (BMI; kg/m2) was measured and adjusted because underweight was suggested to be related with tinnitus.[22] The BMI groups were divided as follows: underweight (<18.5), normal (≥18.5 to <23), and obesity (≥23). The histories of diabetes mellitus, hypertension, depression, and dyslipidemia were considered due to their relations with tinnitus.[122324] The presence of medical histories of diabetes mellitus, hypertension, depression, and dyslipidemia were determined on the basis of diagnoses made by physicians, as reported by the participants. The presence of tinnitus was surveyed using the question “Have you ever heard a sound(s) (buzzing, hissing, ringing, humming, roaring, or machinery noise) in your ear in the past year?”[22] Participants who answered “yes” were included in the tinnitus group and those who answered “no” were included in the no-tinnitus group. The histories of noise exposures were considered due to their contributions to development of tinnitus.[25] Occupational noise exposure was surveyed using the question, “Have you ever worked for 3 or more months in place with loud noise (for instance, machine sound or generator sound)?”. Sudden noise exposure was surveyed using the question, “Have you ever exposed to an explosion sound, such as gun sound or blast?”. The quality of life was surveyed because tinnitus-related distress has been acknowledged to impair the quality of life.[26] EuroQoL (EQ-5D) was examined for the aspects of ambulation, grooming, activities of daily living (ADL), pain or discomfort, and anxiety or depression (www.euroqol.org). The pure-tone air-conduction auditory threshold test was conducted in a sound-proof booth by using an audiometer (GSI SA-203; Entomed Diagnostics AB, Lena Nodin, Sweden).[10] The air-conduction auditory thresholds were evaluated at 0.5, 1, 2, 3, 4, and 6 kHz with Hughson–Westlake method.

Statistical analyses

A Chi-squared test was used to compare the variables between the earphone and control groups. Logistic regression tests with complex sampling were performed for explanatory variables including age, sex, education level, income, BMI, diabetes, hypertension, depression, dyslipidemia, tinnitus, occupational noise exposure, sudden noise exposure, and quality of life to examine which factors were associated with earphone use (dependent variable). In model 1, the covariates of age, sex, income, and education level were adjusted. In model 2, the covariates of age, sex, income, education level, tinnitus, occupational noise exposure, sudden noise exposure, and ADL were adjusted. Two-tailed analyses were conducted, and statistical analyses that yielded a P-value below 0.05 were considered statistically significant. To determine whether age or sex contributed to the differences in the likelihood of tinnitus and anxiety or depression in the earphone group, the samples were divided into three categories based on age: 12- to 20-, 21- to 40-, and >40-year-old males and females. By using the statistical method for the crude data set, the adjusted OR and 95% CI were calculated. The results were statistically analyzed using SPSS (Version 21, IBM).

RESULTS

The 22.9% (449/1955) and 18.1% (355/1955) of earphone users and control participants had tinnitus, respectively (P < 0.001, χ2 = 13.28, Chi-squared test) [Table 1]. The rate of anxiety or depression was higher in the earphone group than in the control group (6.5% [128/1955] vs. 5.0% [97/1955], P = 0.033, χ2 = 4.53, Chi-squared test). The age, education levels, histories of noise exposures, and ADL were different between the earphone and control groups (all P < 0.05). The average age of the participants was 27.42 years old (SD = 13.01) and 27.27 years old (SD = 13.19) in the earphone and control groups (P < 0.001, unpaired t test). The histories of noise exposure were more common in earphone users than control group (8.8% vs. 6.5%, P = 0.005, χ2 = 7.64, Chi-squared test for occupational noise exposure and 20.3% vs. 14.9%, P < 0.001, χ2 = 19.42, Chi-squared test for sudden noise exposure). Other variables including income, BMI, diabetes, hypertension, depression, dyslipidemia, and quality of life variables were comparable between the earphone and control groups. The average hearing thresholds at 500 Hz, 1 kHz, 2 kHz, 3 kHz, 4 kHz, and 6 kHz were comparable between the earphone and control groups [Table 2].

Table 1

General characteristics of earphone user and matched control group

Characteristics	Total participants
	Earphone use (n, %)	Matched control (n, %)	P-value
Number	1955	1955
Age (years old)	27.42	27.27	<0.001
(min–max, SD)	(12–85, 13.01)	(12–85, 13.19)
Sex			0.116
Male	878 (44.8)	927 (47.4)
Female	1077 (55.2)	1028 (52.6)
Education			0.004
No more than primary school	281 (13.9)	344 (17.6)
Lower secondary	310 (15.9)	299 (15.3)
Higher secondary	714 (37.6)	626 (32.0)
Postsecondary	650 (33.2)	686 (35.1)
Income status			0.727
Low	451 (23.1)	448 (22.9)
Low intermediate	471 (24.1)	496 (25.4)
High intermediate	500 (25.6)	475 (24.3)
High	533 (27.3)	536 (27.4)
BMI
Under weight	240 (12.3)	288 (14.7)	0.226
Normal	1280 (65.5)	1270 (65.0)	0.579
Obesity	435 (22.2)	397 (20.3)	0.158
Diabetes mellitus			0.393
Yes	22 (1.1)	28 (1.4)
No	1933 (98.9)	1927 (98.6)
Hypertension			0.735
Yes	75 (3.8)	71 (3.6)
No	1880 (96.2)	1884 (96.4)
Depression			0.420
Yes	35 (1.8)	42 (2.1)
No	1920 (98.2)	1913 (97.9)
Dyslipidemia			0.666
Yes	46 (2.4)	42 (2.1)
No	1909 (97.6)	1913 (97.9)
Tinnitus			<0.001
Yes	449 (22.9)	355 (18.1)
No	1526 (77.1)	1600 (81.9)
Occupational noise exposure			0.005
Yes	173 (8.8)	127 (6.5)
No	1782 (91.2)	1828 (93.5)
Sudden noise exposure			<0.001
Yes	397 (20.3)	292 (14.9)
No	1558 (79.7)	1663 (85.1)
Quality of life
Ambulation			0.439
Yes	59 (3.0)	51 (2.6)
No	1896 (97.0)	1904 (97.4)
Grooming			0.058
Yes	13 (0.7)	5 (0.3)
No	1942 (99.3)	1950 (99.7)
Activities of daily living			0.027
Yes	43 (2.2)	25 (1.3)
No	1912 (97.8)	1930 (98.7)
Pain/discomfort			0.590
Yes	186 (9.5)	196 (10.0)
No	1769 (90.5)	1759 (90.0)
Anxiety/depression			0.033
Yes	128 (6.5)	97 (5.0)
No	1827 (93.5)	1858 (95.0)

The factors taken into consideration are propensity score, age, sex, income, and education level. BMI, body mass index; SD, standard deviation.

Table 2

Pure-tone audiogram by frequency

Characteristics	Right side			Left side
Hz	Earphone use	Control group	P-value	Earphone use		Control group	P-value
500 (dB, mean, SD)	9.59 (10.98)	9.90 (11.13)	0.38	9.68 (9.97)		10.10 (9.71)	0.18
1000 (dB, mean, SD)	6.05 (10.56)	5.95 (10.44)	0.75	4.82 (10.29)		5.14 (9.45)	0.31
2000 (dB, mean, SD)	7.01 (11.52)	6.89 (11.04)	0.74	6.49 (11.42)		6.19 (10.87)	0.41
3000 (dB, mean, SD)	7.08 (12.99)	6.99 (12.77)	0.83	6.50 (13.80)		6.57 (12.85)	0.87
4000 (dB, mean, SD)	8.63 (15.09)	8.66 (15.16)	0.94	8.24 (15.55)		8.46 (14.77)	0.65
6000 (dB, mean, SD)	16.61 (16.56)	17.29 (17.22)	0.20	17.13 (17.58)		17.08 (17.06)	0.93
Average for 500, 1000, 2000, 3000 (dB, mean, SD)	7.43 (10.00)	7.43 (9.83)	0.99	6.87 (10.05)		7.00 (9.25)	0.67

SD, standard deviation; dB, decibel. *Independent t test.

The odds for tinnitus were higher in the earphone group than in the control group (adjusted OR = 1.27, 95% CI = 1.09–1.50, P = 0.003 in model 2) [Table 3]. In subgroup analyses according to age and sex, the 12- to 20-year-old group and >40-year-old group showed higher odds for tinnitus in the earphone group (adjusted OR = 1.28, 95% CI = 1.00–1.65, P = 0.050 for the 12- to 20-year-old group and adjusted OR = 1.77, 95% CI = 1.13–2.77, P = 0.013 for the >40-year-old group) [Table 4]. Male group showed higher odds for tinnitus in the earphone group (adjusted OR = 1.41, 95% CI = 1.10–1.82, P = 0.006 for males and adjusted OR = 1.37, 95% CI = 1.07–1.76, P = 0.013).

Table 3

Crude and adjusted odds ratios (95% confidence interval) of each characteristic for earphone use

Characteristics						Earphone use
			Crude		P-value	Model 1*	P-value	Model 2†	P-value
Age‡		20.8		<0.001		‒0.15	0.721	0.99 (0.99–1.00)	0.159
Sex		1.06 (0.96–1.16)		0.20		0.90 (0.79,1.02)	0.12	1.10 (0.90–1.29)	0.10
Diabetes		0.13 (0.08–0.20)		<0.001§		0.78 (0.42–1.42)	0.47
Hypertension		0.14 (0.11–0.18)		<0.001§		1.05 (0.74–1.49)	0.80
Dyslipidemia		0.20 (0.15–0.27)		<0.001§		1.09 (0.70–1.71)	0.74
Tinnitus		1.09 (0.97–1.21)		0.11		1.33 (1.13–1.56)	<0.001§	1.27 (1.09–1.50)	0.003§
Depression		0.44 (0.30–0.62)		<0.001§		0.83 (0.51–1.33)	0.49
Occupational noise exposure		1.40 (1.10–1.77)		0.006§		1.41 (1.11–1.80)	0.006§	1.32 (1.03–1.69)	0.027§
Sudden noise exposure		1.45 (1.23–1.71)		<0.001§		1.75 (1.44–2.13)	<0.001§	1.70 (1.40,2.06)	<0.001§
Quality of life
Ambulation		0.18 (0.13–0.23)		<0.001§		1.16 (0.78–1.73)	0.49
Grooming		0.16 (0.08–0.28)		<0.001§		2.61 (0.87–9.37)	0.095
ADL		0.23 (0.16–0.31)		<0.001§		1.73 (1.03–2.97)	0.036§	1.56 (0.92–2.62)	0.097
Pain/discomfort		0.37 (0.31–0.43)		<0.001§		0.94 (0.75–1.17)	0.62
Anxiety/depression		0.59 (0.49–0.72)		<0.001§		1.34 (1.01–1.78)	0.039§	1.32 (1.14–1.52)	0.040§

Adjusted model for age, sex, socioeconomic status, and education level. †Adjusted model for age, sex, socioeconomic status, education level, tinnitus, occupational noise exposure, sudden noise exposure, activities of daily living (ADL), and anxiety/depression. ‡β-Coefficient of simple linear regression model. §Fisher exact test model. Significance at P < 0.001.

Table 4

Crude and adjusted odds ratios (95% confidence interval) of tinnitus, depression, quality of life related to anxiety and depressive mood for earphone use by gender

Characteristics	Earphone use					Model 2†	P-value
		Crude	P-value	Model 1*	P-value
Male
Tinnitus		1.02 (0.85–1.22)	0.82	1.41 (1.10–1.82)	0.006‡	1.37 (1.07–1.76)	0.013‡
Occupational noise exposure		0.732 (0.54–0.99)	0.040‡	1.30 (0.96–1.76)	0.091	1.22 (0.90–1.66)	0.202
Sudden noise exposure		1.70 (1.40–2.07)	<0.001‡	1.77 (1.43–2.21)	<0.001‡	1.73 (1.39–2.15)	<0.001‡
Anxiety/depression		0.56 (0.38–0.82)	0.001	1.17 (0.68–2.00)	0.608	1.02 (0.60–1.71)	0.952
Female
Tinnitus		1.16 (1.00–1.34)	0.049	1.27 (1.03–1.56)	0.023‡	1.21 (0.98–1.49)	0.073
Occupational noise exposure		1.58 (1.05–2.37)	0.029‡	1.60 (1.06–2.42)	0.027‡	1.55 (1.02–2.35)	0.040‡
Sudden noise exposure		1.63 (0.92–2.89)	0.097	1.68 (0.94–2.98)	0.080	1.57 (0.88–2.81)	0.128
Anxiety/depression		0.62 (0.49–0.77)	<0.001	1.38 (0.99–1.95)	0.049	1.36 (1.00–1.89)	0.050‡
12–20 years old
Tinnitus		1.55 (1.24–1.96)	<0.001‡	1.35 (1.05–1.75)	0.017‡	1.28 (1.00–1.65)	0.050‡
Occupational noise exposure		1.84 (0.61–5.51)	0.277	1.55 (0.51–4.73)	0.442	1.55 (0.51–4.76)	0.441
Sudden noise exposure		1.31 (0.70–2.44)	0.401	1.40 (0.74–2.63)	0.301	1.33 (0.71–2.52)	0.377
Anxiety/depression		2.23 (0.85–6.03)	0.072	1.53 (0.57–4.35)	0.377	1.04 (0.41–2.63)	0.939
21–40 years old
Tinnitus		1.33 (1.11–1.59)	0.002‡	1.21 (0.96–1.53)	0.104	1.12 (0.89–1.42)	0.335
Occupational noise exposure		1.16 (0.87–1.55)	0.316	1.15 (0.85–1.54)	0.368	1.10 (0.82–1.49)	0.529
Sudden noise exposure		1.63 (1.33–2.00)	<0.001‡	2.18 (1.67–2.84)	<0.001‡	2.15 (1.64–2.80)	<0.001‡
Anxiety/depression		1.27 (0.98–1.65)	0.065	1.43 (1.00–2.05)	0.040‡	1.39 (1.00–1.42)	0.050‡
40+ years old
Tinnitus		1.01 (0.75–1.34)	0.943	1.81 (1.15–2.86)	0.007‡	1.77 (1.13–2.77)	0.013‡
Occupational noise exposure		2.34 (1.41–3.89)	0.001‡	2.44 (1.45–4.10)	0.001‡	2.37 (1.40–3.99)	0.001‡
Sudden noise exposure		1.13 (0.77–1.66)	0.538	1.22 (0.78–1.91)	0.382	1.19 (0.75–1.88)	0.455
Anxiety/depression		0.81 (0.52–1.19)	0.319	1.08 (0.61–1.94)	0.783	1.04 (0.59–1.83)	0.884

*Adjusted model for age, sex, socioeconomic status, and education level. †Adjusted model for age, sex, socioeconomic status, education level, tinnitus, occupational noise exposure, sudden noise exposure, activities of daily living (ADL), and anxiety/depression. ‡Significance at P < 0.05.

For anxiety or depression, the earphone group showed 1.32-times higher rate of anxiety or depression (95% CI = 1.14–1.52, P = 0.040) [Table 3]. According to age groups, the 21- to 40-year-old group demonstrated higher odds for anxiety or depression (adjusted OR = 1.39, 95% CI = 1.00–1.42, P = 0.050) [Table 4]. According to sex, female group showed higher odds for anxiety or depression in the earphone group (adjusted OR = 1.36, 95% CI = 1.00–1.89, P = 0.050).

Earphone users showed higher odds for noise exposures than control group (adjusted OR = 1.32, 95% CI = 1.03–1.69, P = 0.027 for occupational noise exposure and adjusted OR = 1.70, 95% CI = 1.40–2.06, P < 0.001 for sudden noise exposure) [Table 3]. In subgroup analyses, noise exposure was related with earphone use in both the 21- to 40-year-old group and >40-year-old group (adjusted OR = 2.15, 95% CI = 1.64–2.90, P < 0.001 for 21- to 40-year-old group and adjusted OR = 2.37, 95% CI = 1.40–3.99, P = 0.001 for >40-year-old group) [Table 4]. Noise exposure was associated with earphone use in both male and female groups (adjusted OR = 1.73, 95% CI = 1.39–2.15, P < 0.001 for male group and adjusted OR = 1.36, 95% CI = 1.00–1.89, P = 0.050 for female group).

DISCUSSION

Earphone use was associated with tinnitus and anxiety or depression, even though the hearing thresholds were within the normal range. In our study, earphone users showed 1.27-times higher odds for tinnitus and 1.32-times higher odds for anxiety or depression. Male earphone users showed increased odds for tinnitus. According to age, younger (12–20 years old) and middle to older (>40 years old) earphone users had greater incidence of tinnitus. For anxiety or depression, the young adult (21–40 years old) earphone users showed greater incidence of anxiety or depression. These results implied the adverse otologic and psychologic effects of earphone use. Thus, earphone users warrant meticulous clinical attention. Although a few previous studies reported the hazards of abuse of earphones or personal listening devices, this study added to the prior findings by including the results of pure-tone audiometry and by using a representative population cohort.

In line with the present results, prior studies demonstrated a high rate of tinnitus in earphone users.[18] A cross-sectional study on university school students who used earphones reported that approximately 35.7% of students experienced tinnitus and, in particular, 25.9% of students experienced tinnitus after earphone use.[18] A cross-sectional study on 13- to 17-year-old participants reported that the most frequent symptoms after earphone use were hearing loss (21%) and tinnitus (11%). In addition, earphone use in noisy conditions was significantly related to tinnitus (P = 0.003).[27] Another cross-sectional study on high-school students reported that approximately 20.9% of students experienced tinnitus immediately after earphone use.[28] However, these studies included small study populations and lacked the pure-tone audiometry results, and hence, the possible confounders of hearing loss could not be excluded. In addition, most previous studies were performed on adolescent or young adults and did not include unbiased control populations. The positive association with tinnitus was concordant between the general population with normal hearing levels and the control group in this study. According to age groups, the influence of earphone use on tinnitus was high in both the younger and older populations in this study. Although the 21- to 40-year-old group did not show a significant association between earphone use and tinnitus, earphone users in this age group showed increased anxiety or depression. Therefore, the adverse effects of earphone use could not be excluded in this age group.

Tinnitus in earphone users could be associated with hidden or subclinical hearing impairments. Earphone use could impact the auditory function by reducing outer hair-cell function or desynchronizing the auditory nervous pathways. A study on medical students aged 17 to 22 years old showed a significant decrease in distortion-product otoacoustic emissions after 2 hours of earphone use with an average sound pressure of 98.29 dB SPL (P = 0.004), even though their hearing thresholds were ≤15 dB HL.[29] This implied the effects of sound pressure via earphones could impair outer hair-cell functions, which could not be detected using pure-tone audiometry. Patients with tinnitus having normal hearing were suggested to have abnormal activities of the cochlear nucleus or higher-order auditory systems.[3031] Thus, earphone use could impact unspecified auditory nervous system problems that contribute to the development of tinnitus.

Earphone users showed increased rate of anxiety or depression. In addition to the auditory symptoms, tinnitus and related hearing problems could impact anxiety or depression in earphone users. Tinnitus was related to anxiety and depression in several previous studies.[3233] Approximately 45% of patients with tinnitus had anxiety disorder, and several brain areas and corticosubcortical nervous circuits were localized in the causative pathologies.[33] The plausible mechanisms for contribution of tinnitus on anxiety and depression have been suggested, such as tinnitus-related distress and hyperactivation of limbic and sympathetic nervous system.[343536] A cognitive-behavioral model of tinnitus was hypothesized that the importance of interpretations of tinnitus by perceived patients which can provoke emotional distress and maintaining the tinnitus-related distress via feedback loops involving cognitive functions, such as selective attention, monitoring, and counterproductive safety behaviors.[34] For instance, tinnitus distress was correlated with depression and anxiety (β = 0.68 and 0.47, both P < 0.001).[35] To attenuate the potential influence of tinnitus on the association between earphone use and anxiety or depression, the model 2 was adjusted for tinnitus. The results indicated the association of earphone use after adjusted for tinnitus. Therefore, it could be presumed that the association of earphone use with tinnitus was not solely dependent on the presence of tinnitus in earphone users. Indeed, the annoyance of noise itself could contribute to the anxiety or depression. Noise exposure was associated with 1.42-times (95% CI = 1.15–1.74) and 1.20-times (95% CI = 1.00–1.45) higher rates of anxiety and depression, respectively.[37] Similarly, road traffic noise exposure was associated with a 1.17-times higher rate of depression (95% CI = 1.10–1.25).[38] In subgroup analyses, female and 21- to 40-year-old groups demonstrated association of earphone use with anxiety or depression in the present study. This can be partially explained by the relatively high prevalence of anxiety or depression in these sex and age population. The relatively high number of anxiety or depression in these subgroups could potentiate the statistical power for the association between tinnitus and anxiety or depression.

This study was based on a large, representative population. Both demographic and socioeconomic factors were matched between the earphone and control groups. Because the pattern of earphone use is largely determined by these demographic and socioeconomic factors, matching these variables was crucial. In addition, the medical histories were comparable between the earphone and control groups. Hearing loss and tinnitus are known to be related to medical histories, including those of hypertension and diabetes. Therefore, these factors should be considered to minimize the potential confounding effects. Moreover, pure-tone audiometry was conducted in this study, and it confirmed the equivalent hearing levels between the earphone and control groups. The normal hearing level of the earphone group attenuated the possible confounding effects of hearing loss on tinnitus and anxiety or depression. Nevertheless, the present study had a few limitations. Although this cohort had undergone pure-tone audiometry, the hearing thresholds at higher frequencies were not available, which could be affected by noise-induced hearing threshold shifts.[39] Other audiologic tests, such as otoacoustic emissions, electrocochleography, auditory brainstem response, electrophysiologic tests, and speech recognition in noise with and without temporal distortion, could be complementary tests to early detect the noise-induced hearing loss.[3140] Noise exposure via earphones could induce high-frequency hearing loss. In addition, the duration, severity, and pitch of tinnitus were heterogeneous because of the lack of information on the details of tinnitus and tinnitograms. For earphone use, the background noise levels when the participants wore earphones could not be evaluated. Although occupational noise and sudden noise exposures were considered, other noise exposures, such as leisure noise exposure, were not available in the present study. Future studies on the quantitative relationship between earphone abuse and tinnitus and anxiety or depression are warranted to address these questions.

CONCLUSION

Earphone use was related to tinnitus and anxiety or depression. Moreover, a normal hearing level could not guarantee freedom from the adverse effects of earphone use on tinnitus and anxiety or depression.

Financial support and sponsorship

This research was supported by funding from the National Research Foundation (NRF) of Korea (NRF- 2020R1A2C4002594).

Conflicts of interest

There are no conflicts of interest.