Associations Between Tinnitus and Hearing Loss Among Noise-Exposed Workers in the United States From 1999 to 2020: A Cross-Sectional Study

Abstract

Importance

Tinnitus is a prevalent condition among noise-exposed workers, with significant implications for hearing health and quality of life. Identifying associated factors can inform prevention and management strategies.

Objectives

To identify factors associated with tinnitus prevalence among noise-exposed workers and assess the effectiveness of hearing protection in mitigating tinnitus.

Design

Cross-sectional study.

Setting

Data from the National Health and Nutrition Examination Surveys database.

Participants

This study included 4931 noise-exposed workers (mean age 48.9 ± 0.3 years). Participants were categorized into speech-frequency hearing loss (SFHL; n = 1032, mean age 67.2 ± 0.4 years) and high-frequency hearing loss (HFHL; n = 1634, mean age 62.9 ± 0.3 years) groups based on hearing threshold levels.

Intervention or Exposures

Noise exposure duration, hearing loss severity, demographics, and medical history. Hearing protection usage was assessed for its effectiveness in preventing tinnitus.

Main Outcome Measures

Primary outcome: tinnitus prevalence. Associations were analyzed using logistic regression, with factors including self-reported noise exposure, hearing loss severity, demographics, and medical history.

Result

Tinnitus prevalence was 32.85% in the group with SFHL and 29.99% in the group with HFHL. Prolonged noise exposure and greater hearing loss severity were associated with tinnitus in both groups. Hearing protection usage was potentially linked to a lower tinnitus prevalence in HFHL but not in SFHL. For SFHL, univariate analysis showed lower tinnitus prevalence in older age and females, while Caucasian ethnicity and higher income were associated with higher prevalence. Multivariate analysis indicated that older age was positively associated with tinnitus (P < .05). In HFHL, factors such as higher BMI, higher educational level, and elevated Patient Health Questionnaire-9 (PHQ9) scores were significantly associated with tinnitus prevalence.

Conclusions

Prolonged noise exposure and hearing loss severity among noise-exposed workers were associated with tinnitus prevalence in SFHL and HFHL. Hearing protection showed tendency to reduce tinnitus prevalence in HFHL but had a limited effect in SFHL. Higher BMI, higher education levels, and elevated PHQ9 scores were significantly associated with tinnitus in HFHL, warranting further research into protective strategies.

Relevance

Future studies should explore alternative protective strategies for SFHL patients and refine tinnitus prevention approaches in noise-exposed workers.

Graphical Abstract

Keywords

noise exposure hearing loss tinnitus associated factors NHANES

Key Messages

Prolonged work-related noise exposure is a primary factor associated with tinnitus in individuals with hearing loss among noise-exposed workers.

Hearing protection had tendency to reduce tinnitus risk in high-frequency hearing loss cases but not in speech-frequency hearing loss cases.

Higher BMI, higher educational level, and elevated Patient Health Questionnaire-9 scores are significantly associated with an increased odds of tinnitus in individuals with hearing loss among noise-exposed workers, highlighting potential areas for targeted prevention.

Introduction

Noise exposure is one of the leading causes of hearing loss worldwide.¹ The World Health Organization (WHO) estimates that ~5% of the global population suffers from noise-induced hearing loss (NIHL),² with noise exposure accounting for about one-third of all hearing loss causes.³ Particularly, around 16% of disabling hearing loss in adults is attributed to excessive occupational noise.⁴ NIHL significantly diminishes the quality of life of individuals and imposes a substantial economic burden on governments.² Currently, due to the complexity of cochlear damage mechanisms in NIHL, no effective pharmacotherapeutic interventions are available, and personalized prevention is an essential way to reduce the prevalence.^2,5,6

Individuals over-exposed to noise often suffer from high-frequency hearing loss (HFHL), accompanied by bilateral or unilateral tinnitus.^7,8 Tinnitus is usually defined as the conscious perception of sound without an external stimulus⁹ and is essentially an auditory hallucination that can be continuous, intermittent, or manifest as sounds of varying frequencies and tones.¹⁰ It has been reported that tinnitus affects 10% to 15% of people,¹¹ with a prevalence of 14.4% in adults and 13.6% in minors.¹² Studies show that tinnitus occurrence is linked to various factors, the most significant being hearing loss, especially HFHL.¹⁰ This hearing loss can be age-related¹³ or noise-related.⁵ For instance, a retrospective study found that 83% of a cohort of 531 noise-exposed patients experienced HFHL, which was positively correlated with tinnitus loudness.⁷ Similarly, up to 80% of military personnel with NIHL suffer from tinnitus.^7,14,15 However, these findings are challenged by an animal study, which suggested that tinnitus is most prevalent with mild-to-moderate noise exposures and that higher levels of noise exposure do not necessarily lead to greater chronic tinnitus severity.¹⁶ It is important to note that, as an animal study, direct comparisons to human populations are inherently limited. Furthermore, these findings do not necessarily contradict existing human evidence but highlight the complexity of the relationship. To better clarify the relationship between hearing loss, tinnitus prevalence, and severity, further studies involving larger human populations are needed.

Besides the degree of hearing loss, other factors such as obesity, smoking, alcohol consumption, hypertension, previous head injuries, and genetics are considered potential risk factors for the prevalence of tinnitus.^17,18 Studies have also found that smoking,^19,20 hypertension,²¹ genetics,²² sex,²³ and diabetes²⁴ are highly associated with the severity of NIHL. Although previous studies have identified similarities in the risk factors for tinnitus and NIHL, differences have also been observed. For example, a recent nationwide cohort study in Denmark reported that road traffic noise might increase the risk of tinnitus.²⁵ However, as this study focused on traffic noise in the Danish population, it highlights the need for further investigations into the risk factors for tinnitus with occupational noise-related hearing loss, particularly across diverse ethnicities.

The current study aims to investigate the relationship between hearing loss, along with other factors such as age, gender, BMI, educational level, and mental health with the prevalence and severity of tinnitus in noise-exposed workers with hearing loss, using publicly available data from the National Health and Nutrition Examination Surveys (NHANES). The findings seek to draw attention to the impact of NIHL on tinnitus and underline the importance of preventing hazardous noise exposures in the workplace to mitigate further auditory damage.

Methodology

Study Population

The NHANES is a cross-sectional study assessing the health and nutrition of U.S. adults and children.^26,27 This analysis used data from the 1999 to 2004, 2011 to 2012, and 2015 to 2020 NHANES cycles, selected for their inclusion of work-related noise exposure, audiometric measurements, and tinnitus prevalence. Participants who were unable to remove their hearing aids for testing (eg, cochlear implants or bone-anchored hearing aids) or those experiencing severe ear pain during exams were excluded. The final sample was of 4931 participants: 1032 with speech-frequency hearing loss (SFHL) and 1632 with HFHL. Bilateral hearing loss criteria were used to categorize participants. SFHL was defined as a pure-tone average (PTA) >20 dB hearing level (HL) at 0.5, 1, 2, and 4 kHz in the better ear, while HFHL was identified as a PTA >20 dB HL at 3, 4, 6, and 8 kHz. Individuals with middle ear diseases were excluded during data collection to minimize the impact of conductive hearing loss.

Noise Exposure Among Workers

Work-related noise exposure in NHANES was defined as exposure to loud noise at work for at least 4 hours/day over a minimum period of 3 months.^28,29 Definitions varied slightly across cycles, but consistently focused on exposure requiring raised voices to communicate. We also collected data on the frequency of work-related noise exposure. For detailed definitions across cycles, see the Methods section in Supplemental Materials for more information.

Tinnitus

Tinnitus was defined based on survey questions from NHANES that identify participants reporting ringing, roaring, or buzzing lasting ≥5 minutes in the past 12 months. For the 1999 to 2004 cycle, tinnitus was assessed using the question: “In the past 12 months, have you ever had ringing, roaring, or buzzing in your ears?” Participants who responded “yes” were classified as having tinnitus. For the 2011 to 2012 and 2015 to 2020 cycles, tinnitus was assessed using the question: “In the past 12 months, have you been bothered by ringing, roaring, or buzzing in your ears or head that lasts for 5 minutes or more?” Similarly, participants who responded “yes” were classified as having tinnitus. Due to differences in the NHANES questionnaires across data collection periods, the definition of tinnitus varied, with the 1999 to 2004 cycle using a broader definition and the 2011 to 2012 and 2015 to 2020 cycles focusing on tinnitus that was perceived as bothersome. To address this, data from all cycles were combined, and a weighted analysis was employed to minimize the potential bias and ensure the robustness of findings.

The frequency of tinnitus was assessed using questions that asked participants to report how often they experienced these symptoms. For 1999 to 2004, responses included options such as “Less frequently than once a month,” “At least once a month,” “At least once a week,” “At least once a day,” and “Almost always.” A similar format was used for the 2011 to 2012 and 2015 to 2020 data, asking participants to indicate the frequency of their tinnitus symptoms over the past 12 months.

Audiometric Assessment

Audiometric examinations were conducted in sound-isolating rooms within Mobile Examination Centers (MECs) by trained examiners. Air conduction hearing threshold tests were performed for both ears across 7 frequencies (0.5, 1, 2, 3, 4, 6, and 8 kHz), with thresholds ranging from −10 to 120 dB HL. NHANES follows a standardized protocol for these examinations, detailed information for which is available on their website (https://wwwn.cdc.gov/Nchs/Nhanes/). Hearing levels were categorized according to the 2021 WHO’s guidelines: normal (<20 dB), mild (20-35 dB), moderate (35-50 dB), moderate-to-severe (50-65 dB), severe (65-80 dB), profound (80-95 dB), and deafness (>95 dB). In this study, a threshold above 50 dB was defined as “moderate-severe or greater,” due to the limited sample size of participants with severe to profound hearing loss and those with deafness.

Demographics and Hearing-Related Variables

Data on demographics and hearing-related variables were presented in the Methods section in the Supplemental Material.

Statistical Analysis

NHANES sample weights were applied in the statistical analysis to account for complex sampling and non-response, conducted in R (version 4.1.3; R Foundation for Statistical Computing, Vienna, Austria). Continuous variables were presented as mean ± standard error of the mean, and categorical variables as percentages with 95% confidence intervals (CIs). Student’s t-tests and Pearson’s chi-square tests were used to compare continuous and categorical variables, respectively, between participants with and without tinnitus in the SFHL and HFHL groups among noise-exposed workers. Variables examined included demographic characteristics (eg, age, sex, BMI, ethnicity), audiometric thresholds, and noise exposure duration. Univariate logistic regression was performed to calculate odds ratios (ORs) for associations between hearing loss and tinnitus. Models were progressively adjusted for confounders, including demographic factors, degree of hearing loss, comorbidities (eg, BMI, hypertension, diabetes), and lifestyle factors (eg, smoking and alcohol use). This stepwise approach allowed us to evaluate the independent effects of noise exposure on tinnitus while minimizing confounding bias. To address potential overestimation of effect sizes by ORs in logistic regression, we utilized log-binomial and Poisson regression models to directly calculate adjusted relative risks (RRs). These approaches are particularly suited for our study, where the outcome (tinnitus and hearing loss) was not rare, ensuring that risk estimates remained interpretable and clinically relevant. All models were adjusted for covariates, including age, sex, occupational noise exposure duration, and comorbidities. The analysis was adjusted for potential confounders. Multiple comparisons were conducted across these variables and groups, with adjustments made to address the increased likelihood of Type I errors. To reduce the probability of Type I errors, the Bonferroni correction was applied. The significance threshold was set at P < .05 for individual tests, ensuring a conservative approach to statistical inference.

Results

Characteristics of Hearing Loss Among Noise-Exposed Workers With and Without Tinnitus

Our study encompassed a comprehensive analysis of 4931 subjects, comprising 1032 individuals with SFHL, 32.85% of whom experienced tinnitus, and 1634 with HFHL, with 29.99% reporting tinnitus. Only 13 participants experienced both SFHL and HFHL concurrently, indicating that these 2 subgroups were largely independent of each other. Detailed descriptions of tinnitus characteristics and noise exposure for both SFHL and HFHL cohorts are illustrated in Figures S1 and S2, respectively. Baseline demographics and clinical characteristics are systematically summarized in Table 1 include a summary of hearing loss severity. The mean ages for SFHL subjects with and without tinnitus were 60.8 ± 0.9 and 63.6 ± 0.7 years, respectively. It was noted that the level of hearing loss did not significantly differ between SFHL subjects with and without tinnitus. Tinnitus was predominantly observed in younger individuals, females, White ethnicity, those with more income, and those exposed to noise for extended durations. In the HFHL cohort, average ages were 59.1 ± 0.5 years (without tinnitus) and 58.2 ± 0.8 years (with tinnitus). Subjects with tinnitus exhibited more pronounced HFHL, increased Patient Health Questionnaire-9 (PHQ9) scores, and longer exposure to noise. No significant differences were observed in the prevalence of physical illnesses such as diabetes mellitus, hypertension, or cardiovascular disease among participants with or without tinnitus in both SFHL and HFHL groups.

Table 1.

Demographic and Clinical Characteristics of the Participants.

Variable	Speech-frequency hearing loss (n = 1032)^a			High-frequency hearing loss (n = 1634)^b
	Without tinnitus	With tinnitus	P	Without tinnitus	With tinnitus	P
	(n = 339)	(n = 693)	P	(n = 490)	(n = 1144)	P
Age (years)	63.6 ± 0.7	60.8 ± 0.9	.02	59.1 ± 0.5	58.2 ± 0.8	.3
BMI (kg/m²)	29.91 ± 0.33	30.19 ± 0.57	.72	29.67 ± 0.30	30.59 ± 0.37	.02
Right ear speech-frequency hearing (dB)	34.71 ± 0.82	35.00 ± 1.01	.78	25.37 ± 0.60	27.66 ± 0.94	.01
Left ear speech-frequency hearing (dB)	35.90 ± 0.81	36.18 ± 0.97	.78	26.29 ± 0.58	29.53 ± 0.85	<.001
Right ear high-frequency hearing (dB)	54.98 ± 0.98	58.84 ± 1.48	.01	44.07 ± 0.80	50.25 ± 1.24	<.0001
Left ear high-frequency hearing (dB)	57.54 ± 1.07	60.80 ± 1.42	.04	46.58 ± 0.82	53.24 ± 0.97	<.0001
Sex (%)			<.001			.07
Male	146 (77.94)	45 (22.06)		882 (66.05)	399 (33.95)
Female	547 (60.96)	294 (39.04)		262 (72.54)	91 (27.46)
Ethnicity (%)			<.001			<.0001
White	370 (61.39)	194 (38.61)		524 (64.66)	267 (35.34)
Black	117 (75.31)	42 (24.69)		237 (78.67)	69 (21.33)
Mexican	79 (62.54)	50 (37.46)		167 (71.54)	71 (28.46)
Other	127 (77.40)	53 (22.60)		216 (76.47)	83 (23.53)
Education (%)			.31			.14
Less than a high school graduate	86 (71.07)	42 (28.93)		153 (76.56)	57 (23.44)
High school graduate	343 (66.27)	156 (33.73)		534 (68.95)	220 (31.05)
Some college or more	264 (60.92)	141 (39.08)		457 (64.83)	213 (35.17)
Income			.01			.39
<$20,000	130 (73.84)	52 (26.16)		242 (70.53)	91 (29.47)
Over $20,000	306 (60.47)	171 (39.53)		582 (66.04)	266 (33.96)
Don’t know/refused	257 (69.10)	116 (30.90)		320 (70.30)	133 (29.70)
DM			.8			.92
Normal	362 (63.98)	178 (36.02)		653 (68.26)	274 (31.74)
IFG	50 (64.09)	22 (35.91)		81 (66.57)	32 (33.43)
IGT	10 (52.21)	8 (47.79)		26 (63.50)	14 (36.50)
DM	271 (65.48)	131 (34.52)		384 (66.05)	170 (33.95)
Hypertension			.62			.28
No	228 (62.64)	128 (37.36)		442 (69.36)	181 (30.64)
Yes	463 (64.81)	211 (35.19)		700 (65.87)	309 (34.13)
CVD			.74			.49
No	472 (63.55)	230 (36.45)		869 (67.86)	352 (32.14)
Yes	219 (65.19)	109 (34.81)		275 (65.44)	138 (34.56)
Hyperlipidemia			.23			.47
No	110 (56.98)	64 (43.02)		235 (70.03)	84 (29.97)
Yes	583 (65.38)	275 (34.62)		909 (66.76)	406 (33.24)
Stroke			.75			.9
No	624 (64.33)	300 (35.67)		1046 (67.48)	441 (32.52)
Yes	65 (61.23)	35 (38.77)		96 (66.60)	45 (33.40)
Cancer			.89			.91
No	549 (63.84)	251 (36.16)		957 (67.45)	381 (32.55)
Yes	142 (64.57)	87 (35.43)		187 (66.98)	108 (33.02)
Smoke			.3			.66
Never	238 (62.56)	116 (37.44)		436 (67.92)	171 (32.08)
Former	321 (67.72)	153 (32.28)		441 (68.65)	205 (31.35)
Current	134 (59.53)	70 (40.47)		267 (64.67)	114 (35.33)
Alcohol			.13			.2
Never	59 (67.53)	22 (32.47)		108 (68.85)	32 (31.15)
Former	80 (60.73)	36 (39.27)		170 (70.67)	60 (29.33)
Mild	212 (59.79)	127 (40.21)		347 (63.53)	171 (36.47)
Moderate	63 (56.51)	37 (43.49)		109 (61.28)	63 (38.72)
Heavy	97 (75.80)	43 (24.20)		192 (73.72)	78 (26.28)
PHQ9^c			.69			.01
(0, 9)	582 (64.55)	267 (35.45)		944 (68.41)	378 (31.59)
(10, 27)	51 (61.70)	45 (38.30)		81 (53.67)	72 (46.33)
Time exposure			.03			.05
<3 Months	25 (88.88)	5 (11.12)		44 (91.35)	10 (8.65)
3-11 Months	37 (73.02)	19 (26.98)		64 (73.93)	25 (26.07)
1-2 Years	63 (67.70)	32 (32.30)		113 (69.18)	52 (30.82)
3-4 Years	41 (46.78)	30 (53.22)		83 (65.15)	42 (34.85)
5-9 Years	105 (65.15)	44 (34.85)		179 (71.07)	64 (28.93)
10-14 Years	83 (73.62)	34 (26.38)		132 (65.13)	52 (34.87)
15 or more years	303 (61.20)	159 (38.80)		453 (63.36)	220 (36.64)

Abbreviations: BMI, body mass index; CVD, cardiovascular disease; DM, diabetes mellitus; IFG, impaired fasting glycemia; IGT, impaired glucose tolerance; PHQ9, Patient Health Questionnaire-9; PTA, pure-tone average; SEM, standard error of the mean.

Speech-frequency hearing loss was defined as a PTA hearing threshold exceeding 20 dB in the better ear at speech frequencies (0.5, 1, 2, and 4 kHz). The number of speech-frequency hearing loss was 1015. Continuous variable uses mean ± SEM. Classification variables are used in numbers or percentages (%).

High-frequency hearing loss was defined as a PTA hearing threshold exceeding 20 dB in the better ear at high frequencies (3, 4, 6, and 8 kHz). The number of high-frequency hearing loss was 1843. Continuous variable uses mean ± SEM. Classification variables are used in numbers or percentages (%).

PHQ9: (0-9): normal; (10-27): depression.

Association Between Tinnitus and Hearing Loss Among Noise-Exposed Workers: Results From Univariate Logistic Regression

Figure 1 delineates the univariate logistic regression analysis outcomes that the factors associated with tinnitus in participants with SFHL or HFHL, utilizing MEC sample weights.

Figure 1.

Univariate logistic regression analyses on the association between various health and audiological factors, clinical factors, and tinnitus according to type of hearing loss (SFHL and HFHL). (a) PTA of speech frequency in right ear, (b) PTA of speech frequency in left ear, (c) PTA of high frequency in right ear, and (d) PTA of high frequency in left ear. In the statistical analysis, this group was used as the reference for comparison (*). BMI, body mass index; DM, diabetes mellitus; IFG, impaired fasting glycemia; GT, impaired glucose tolerance; CVD, cardiovascular disease; PHQ9, Patient Health Questionnaire-9; Ref, reference; PTA, pure-tone average; LHH, Left ear high-frequency hearing; LSH, Left ear speech-frequency hearing; RHH, Right ear high-frequency hearing; RSH, Right ear speech-frequency hearing.

In the SFHL group, advanced age (OR = 0.98, 95% CI: 0.97-1.00, P < .05) and female gender (OR = 0.44, 95% CI: 0.29-0.68, P < .001) were associated with a lower odds of tinnitus. Compared to White individuals, Black (OR = 0.52, 95% CI: 0.32-0.86, P = .01) and other ethnicities (OR = 0.46, 95% CI: 0.29-0.73, P = .001) demonstrated a lower prevalence of tinnitus. Extended noise exposure and higher annual income ($20,000 and above, OR = 1.85, 95% CI: 1.11-3.06, P = .02) were associated with a higher odds of tinnitus. No significant associations were noted in hearing loss levels, educational background, smoking, alcohol consumption, PHQ9 scores, or physical illnesses, including diabetes mellitus, hypertension, hyperlipidemia, cardiovascular disease, and stroke.

For participants with HFHL, similar to SFHL, extended noise exposure time was significantly associated with a higher odds of tinnitus, while non-White individuals (OR = 0.50, 95% CI: 0.34-0.73, P < .001 for Black; OR = 0.56, 95% CI: 0.38-0.82, P < .001 for other ethnicities) were less likely to report tinnitus. Additional factors associated with tinnitus in HFHL, distinct from SFHL, included higher BMI (OR = 1.02, 95% CI: 1.00-1.04, P = .02), more severe HFHL (OR = 1.02, 95% CI: 1.01-1.02, P < .0001), higher educational levels (OR = 1.47, 95% CI: 1.00-2.15, P = .05 for high school graduates; OR = 1.77, 95% CI: 1.16-2.71, P = .01 for college or higher), and elevated PHQ9 scores (OR = 1.87, 95% CI: 1.16-3.01, P = .01). None of the physical illness conditions indicated in this study showed a significant association with the prevalence of tinnitus in the HFHL. In addition, no significant differences were found in age, gender, or annual income among SFHL subjects with and without tinnitus.

In this cohort, an association between the length of work-related noise exposure and tinnitus occurrence was observed. After adjusting for covariates, univariate regression analysis reassessed the association between noise exposure duration and tinnitus prevalence in individuals with hearing loss among noise-exposed workers, as delineated in Table 2. Adjustments included age, sex, ethnicity, education, hearing loss severity, BMI, hypertension, diabetes, smoking, and alcohol consumption. It was discerned that in subjects afflicted with SFHL, the association between noise exposure duration and tinnitus onset was less pronounced. Nonetheless, a duration of noise exposure exceeding 3 years remained significantly associated with a higher odds of tinnitus development (OR = 11.72, 95% CI: 2.52-54.62, P < .01). Conversely, among participants with HFHL, the odds of tinnitus were found to escalate with longer duration of noise exposure, even after adjusting for the aforementioned covariates, as indicated in Table 2.

Table 2.

Univariate and Multivariate Logistic Regression Analysis of Noise Exposure Duration and Tinnitus.

Character	Crude model			Model 1			Model 2			Model 3			Model 4
Character	Estimate	OR (95% CI)	P value	Estimate	OR (95% CI)	P value	Estimate	OR (95% CI)	P value	Estimate	OR (95% CI)	P value	Estimate	OR (95% CI)	P value
Speech-frequency hearing loss
<3 Months	Ref^a	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref
3-11 Months	1.08	2.95 (0.72-12.10)	.13	0.88	2.41 (0.48-12.08)	.28	0.92	2.50 (0.52-12.13)	.25	0.91	2.49 (0.51-12.07)	.25	0.21	1.23 (0.21-7.14)	.81
1-2 Years	1.34	3.81 (1.12-13.00)	.03	1.35	3.85 (0.92-16.07)	.06	1.37	3.93 (0.99-15.66)	.05	1.27	3.57 (0.87-14.70)	.08	1.06	2.89 (0.57-14.60)	.19
3-4 Years	2.21	9.09 (2.51-32.99)	.001	2.39	10.92 (2.64-45.10)	.001	2.4	11.04 (2.72-44.86)	.001	2.38	10.83 (2.72-43.06)	.001	2.46	11.72 (2.52-54.62)	.003
5-9 Years	1.45	4.28 (1.28-14.27)	.02	1.49	4.44 (1.13-17.42)	.03	1.49	4.45 (1.16-17.09)	.03	1.46	4.30 (1.12-16.48)	.03	1.22	3.40 (0.85-13.55)	.08
10-14 Years	1.05	2.87 (0.76-10.78)	.12	1.22	3.40 (0.76-15.21)	.11	1.17	3.22 (0.74-14.04)	.12	1.03	2.81 (0.66-11.99)	.16	1.12	3.07 (0.61-15.48)	.17
15 or more years	1.62	5.07 (1.56-16.47)	.01	1.7	5.49 (1.44-20.87)	.01	1.72	5.60 (1.52-20.63)	.01	1.67	5.32 (1.42-19.91)	.01	1.53	4.62 (1.12-19.03)	.03
High-frequency hearing loss
<3 Months	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref
3-11 Months	1.31	3.72 (1.45-9.55)	.01	1.28	3.58 (1.31-9.80)	.01	1.38	3.97 (1.44-10.99)	.01	1.42	4.13 (1.49-11.39)	.01	1.06	2.88 (0.92-9.06)	.07
1-2 Years	1.55	4.71 (2.21-10.01)	<.001	1.53	4.61 (2.03-10.46)	<.001	1.51	4.54 (2.02-10.20)	<.001	1.52	4.59 (2.01-10.51)	<.001	1.65	5.19 (2.05-13.10)	<.001
3-4 Years	1.73	5.65 (2.41-13.26)	<.001	1.75	5.75 (2.34-14.11)	<.001	1.75	5.74 (2.40-13.74)	<.001	1.7	5.48 (2.24-13.43)	<.001	1.84	6.32 (2.34-17.05)	<.001
5-9 Years	1.46	4.30 (1.81-10.19)	.001	1.52	4.55 (1.82-11.41)	.002	1.53	4.62 (1.91-11.16)	.001	1.52	4.58 (1.87-11.24)	.001	1.71	5.52 (1.97-15.44)	.002
10-14 Years	1.73	5.65 (2.14-14.94)	<.001	1.77	5.88 (2.14-16.13)	<.001	1.78	5.96 (2.20-16.14)	<.001	1.75	5.77 (2.08-15.97)	.001	2.05	7.79 (2.60-23.36)	<.001
15 or more years	1.81	6.11 (2.95-12.64)	<.0001	1.88	6.56 (3.00-14.37)	<.0001	1.86	6.41 (2.99-13.74)	<.0001	1.83	6.26 (2.88-13.64)	<.0001	1.98	7.22 (3.15-16.51)	<.0001

Crude model: The relationship between noise exposure duration and tinnitus. Model 1: adjusted age, sex, ethnicity, and education. Model 2: adjusted age, sex, ethnicity, education, and degree of hearing loss. Model 3: adjusted age, sex, ethnicity, education, degree of hearing loss, BMI, hypertension, and DM. Model 4: adjusted age, sex, ethnicity, education, degree of hearing loss, BMI, hypertension, DM, smoke statute, and alcohol use statute.

Abbreviations: BMI, body mass index; CI, confidence interval; DM, diabetes mellitus; ORs, odds ratios; ref, reference.

In the statistical analysis, this group was used as the reference for comparison.

Multivariate Logistic Regression Analysis of Tinnitus and Associated Factors in Hearing Loss Among Noise-Exposed Workers

Subsequent analysis through multivariate logistic regression was employed to elucidate the influencing factors for tinnitus. A statistical significance threshold of P < .05 was applied for inferential statistics, and a P-value threshold of ≤.2³⁰ was used in univariate analysis for the inclusion of potential associations into the multivariate model. The model also incorporated common vascular risk factors such as hypertension, diabetes, smoking, and alcohol use.¹⁰ Figure 2 presents the results of these multivariate regression analyses. It was noticed that in participants with SFHL, older age (RR = 0.99, 95% CI: 0.98-0.99, P = .01) was associated with a lower odds of tinnitus. Interestingly, we also found that heavy alcohol intake (RR = 0.58, 95% CI: 0.41-0.82, P = .012) was associated with a lower odds of tinnitus. By contrast, moderate hearing loss (RR = 1.32, 95% CI: 1.01-1.73, P = .046) was associated with a higher prevalence of tinnitus. For participants with HFHL, older age (RR = 0.98, 95% CI: 0.97-0.99, P < .001) was similarly associated with a lower odds of tinnitus. Conversely, increased BMI (RR = 1, 95% CI: 0.99-1.02, P = .377), higher educational level (RR = 1.69, 95% CI: 1.13-2.54, P = .012), and elevated PHQ9 scores (RR = 1.47, 95% CI: 1.19-1.83, P < .001) were associated with a higher prevalence of tinnitus. Although there was no statistically significant difference in BMI. It is noteworthy that a history of noise exposure remained significantly associated with tinnitus among individuals with either SFHL or HFHL.

Figure 2.

Multivariate logistic regression analyses on the association between various health and audiological factors, clinical factors, and tinnitus according to type of hearing loss (SFHL and HFHL). In the statistical analysis, this group was used as the reference for comparison (*). BMI, body mass index; DM, diabetes mellitus; IFG, impaired fasting glycemia; IGT, impaired glucose tolerance; CVD, cardiovascular disease; PHQ9, Patient Health Questionnaire-9; ref, reference; SFHL, speech-frequency hearing loss; HFHL, high-frequency hearing loss.

Role of Hearing Protection in Mitigating Tinnitus Among Individuals With Hearing Loss

Noise exposure was consistently associated with a higher odds of tinnitus in both SFHL and HFHL. This finding was observed through univariate and multivariate logistic regression analyses. Based on this, the study further examined the efficacy of hearing protection in reducing tinnitus prevalence. Hearing protection is a widely used strategy for preventing noise-induced hearing injury. The findings indicated that hearing protection had higher tendency to prevent work-related HFHL (RR = 1.12, 95% CI: 0.93-1.35, P = .228) than SFHL (RR = 0.77, 95% CI: 0.54-1.11, P = .154) (Table 3), although there was no statistically significant difference. However, it did not reduce the occurrence of tinnitus in either SFHL or HFHL cohorts (Table 4).

Table 3.

The Association Between the Use of Hearing Protection and Degree of Hearing Loss.

Character	Model^a
Character	RR (95% CI)	P
Speech-frequency hearing loss
Wore devices	Ref^b	Ref
Did not wear devices	0.77 (0.54-1.11)	.154
High-frequency hearing loss
Wore devices	Ref	Ref
Did not wear devices	1.12 (0.93-1.35)	.228

Abbreviations: BMI, body mass index; CI, confidence interval; DM, diabetes mellitus; Ref, reference; RR, relative risk.

The model adjusted noise exposure time, age, sex, ethnicity, education, BMI, hypertension, DM, smoke statute, and alcohol use statute.

In the statistical analysis, this group was used as the reference for comparison.

Table 4.

The Association Between the Use of Hearing Protection and Tinnitus Among Hearing Loss.

Character	Model^a
Character	RR (95% CI)	P
Speech-frequency hearing loss
Wore devices	Ref^b	Ref
Did not wear devices	0.84 (0.62-1.14)	.255
High-frequency hearing loss
Wore devices	Ref	Ref
Did not wear devices	0.95 (0.71-1.26)	.696

Abbreviation: BMI, body mass index; CI, confidence interval; DM, diabetes mellitus; Ref, reference; RR, relative risk.

The model adjusted noise exposure time, age, sex, ethnicity, education, degree of hearing loss, BMI, hypertension, DM, smoke statute, and alcohol use statute.

In the statistical analysis, this group was used as the reference for comparison.

Discussion

In this study, we found that both duration of noise exposure and the severity of hearing loss were significantly associated with tinnitus prevalence among individuals working in noise-exposed environments, across both SFHL and HFHL groups. While demographic and medical history factors were linked to tinnitus, the nature of these associations differed between SFHL and HFHL. The use of hearing protection was associated with a reduced odds of HFHL but showed no significant effect on SFHL. Tinnitus exhibited distinct cardiovascular comorbidity patterns across hearing loss subgroups: hypertension was associated in the group with HFHL, while diabetes demonstrated a significant association with tinnitus in the group with SFHL. This study is the first to comprehensively analyze factors associated with noise-exposed SFHL and HFHL in a large population.

NIHL can result from occupational, recreational, and traffic-related noise. NHANES provided data on occupational and recreational exposure but lacked traffic-related NIHL data. Detailed exposure information was available only for occupational settings, so our study focused on factors associated with occupational NIHL, excluding recreational and traffic-related factors. Given that NIHL is often associated with HFHL,¹⁰ particularly in its initial stages, our analysis encompassed both HFHL and SFHL. The univariate and multivariate logistic regression analyses consistently showed that the duration of noise exposure was significantly associated with an increased odds of tinnitus in cases of noise-exposed HFHL. However, a tendency towards reduction in tinnitus prevalence after hearing protection was observed only among HFHL but not SFHL participants. This phenomenon might be attributed to the fact that excessive noise exposure can cause auditory system damage, leading to permanent hearing loss and tinnitus, with a higher severity observed at high frequencies.^6,25,31 Previous epidemiological studies have reported an increased risk of tinnitus with repeated occupational and leisure noise exposure.^32,33 For instance, a recent nationwide cohort study from Denmark indicated that residential exposure to road traffic noise might elevate the risk of tinnitus.²⁵ Hence, our findings align with these previous results, underscoring the importance of hearing protection.

Additional factors related to demographics and medical history were also examined. In SFHL participants, univariate logistic regression analysis revealed that older age and female gender were associated with a lower odds of tinnitus, whereas White ethnicity and higher annual income were linked with a higher odds. Participants with higher income levels may have better access to healthcare, resulting in more frequent diagnoses and potentially contributing to a higher reported prevalence of tinnitus. By contrast, younger participants generally have lower income levels, which may be associated with lower reports of tinnitus. This could possibly explain why higher annual income is associated with an increased odds of tinnitus. Multivariate analysis showed that older age was associated with a lower odds of tinnitus. This finding, particularly regarding older age as a protective factor, contrasts with existing research of Lisa et al,³⁴ indicating a need for further studies to validate these factors. Based on the age distribution of individuals with tinnitus, as well as univariate and multivariate regression analyses, we observed that older age was associated with lower odds of tinnitus. However, data from a cross-sectional study indicate that the prevalence of hearing impairment is related to age, but this age-related trend was not observed in tinnitus.³⁵ This discrepancy may be attributed to the relatively older age range of our study population (mean age 58.2-63.6 years), where participants might already have noise-induced central auditory changes,¹³ potentially reducing additional tinnitus risk. The observed protective association between alcohol consumption and tinnitus may be linked to its impact on psychological factors.¹⁰ Given the strong psychological component of tinnitus, including its association with stress and anxiety, moderate alcohol use might indirectly alleviate tinnitus symptoms by reducing psychological distress. It is consistent with the PHQ9 results in our study, which highlight the significant role of psychological well-being in tinnitus prevalence. In HFHL participants, univariate analysis showed associations between White ethnicity, higher BMI, higher education level, and elevated PHQ9 score with an increased odds of tinnitus. These associations remained significant in multivariate analysis except for White ethnicity. The consistency in findings between the univariate and multivariate analyses suggests that HFHL more accurately reflects the cochlear damage induced by noise. Interestingly, the aforementioned Danish cohort study²⁵ reported different risk estimates, highlighting differences in the impact of occupational noise in the American population compared to traffic noise in Denmark, particularly in the context of HFHL. Notably, common diseases such as diabetes, hypertension, and cardiovascular conditions did not demonstrate a compounded effect on tinnitus prevalence among NIHL sufferers, though they were potential risk factors for both tinnitus^17,18 and the severity of hearing loss.^21,24

Limitations include reliance on self-reported data for noise exposure and disease definitions, which may affect reliability. It is worth mentioning that the variation in tinnitus definitions between NHANES cycles, in which 1999 to 2004 asked about any tinnitus occurrence, while 2015 to 2020 focused on distress caused by tinnitus, potentially affecting prevalence rates and the overall assessment of tinnitus burden in the population. Also, defining hypertension as “told by a doctor” or “using medication” may affect classification reliability due to non-adherence or lifestyle changes, suggesting the need for direct blood pressure measurements in future studies. In addition, while previous studies^29,36 on tinnitus and hearing loss commonly use ORs, we acknowledge that ORs can overestimate associations in high-prevalence conditions like tinnitus. In our study, we employed log-binomial and Poisson regression models to directly estimate adjusted RRs within a multivariable analytical framework, offering an alternative to traditional ORs derived from logistic regression. Finally, the database does not specify tinnitus laterality, and asymmetric hearing loss was not explicitly analyzed in this study.

Conclusion

In conclusion, our study indicates that noise exposure and hearing loss severity are significantly associated with tinnitus in workers with noise-induced SFHL and HFHL. Hearing protection was potentially linked to reduced tinnitus prevalence only in HFHL. Higher BMI, higher educational level, and elevated PHQ9 scores were also associated with tinnitus in noise-induced HFHL among workers. These findings highlight factors associated with tinnitus in NIHL and suggest potential preventive strategies.

Supplemental Material

sj-docx-1-ohn-10.1177_19160216251347597 – Supplemental material for Associations Between Tinnitus and Hearing Loss Among Noise-Exposed Workers in the United States From 1999 to 2020: A Cross-Sectional Study

Supplemental material, sj-docx-1-ohn-10.1177_19160216251347597 for Associations Between Tinnitus and Hearing Loss Among Noise-Exposed Workers in the United States From 1999 to 2020: A Cross-Sectional Study by Shuo-Wei Yang, Wei Xu, Lin Chen and Shu-Bin Fang in Journal of Otolaryngology - Head & Neck Surgery

Footnotes

Acknowledgements

Thanks to Zhang Jing (Second Department of Infectious Disease, Shanghai Fifth People’s Hospital, Fudan University) for his work on the NHANES database. His outstanding work, nhanesR package and webpage, makes it easier for us to explore the NHANES database.

Author Contributions

S.-W.Y. helped in the analysis of data and manuscript writing; W.X. and L.C. helped in the analysis of data. S.-B.F. helped in concept and design, data analysis, manuscript writing, and final approval of the manuscript.

Consent to Participate

Not applicable.

Consent for Publication

Consent to publish was obtained.

Data Availability Statement

The datasets generated and/or analyzed during the current study are available in the NHANES repository ().

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grants from National Natural Science Foundation of China (no. 82101185), Natural Science Foundation of Guangdong Province (grant no. 2024A1515010530), and Basic Research Program of Guangzhou Municipal Science and Technology Bureau (no. 2024A04J4695).

Ethical Considerations

The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). All information from the NHANES program is available and free for public, so the agreement of the medical ethics committee board was not necessary.

ORCID iDs

Shuo-Wei Yang

Shu-Bin Fang

Supplemental Material

Additional supporting information is available in the online version of the article.

References

World Health Organization. World Report on Hearing. WHO; 2021.

Natarajan

Batts

Stankovic

KM.

Noise-induced hearing loss. J Clin Med. 2023;12(6):2347.

Oishi

Schacht

Emerging treatments for noise-induced hearing loss. Expert Opin Emerg Drugs. 2011;16(2):235-245.

Nelson

Concha-Barrientos

Fingerhut

The global burden of occupational noise-induced hearing loss. Am J Ind Med. 2005;48(6):446-458.

Basner

Babisch

Davis

, et al. Auditory and non-auditory effects of noise on health. Lancet. 2014;383(9925):1325-1332.

Shore

Mechanisms of noise-induced tinnitus: insights from cellular studies. Neuron. 2019;103(1):8-20.

Mazurek

Olze

Haupt

Szczepek

AJ.

The more the worse: the grade of noise-induced hearing loss associates with the severity of tinnitus. Int J Environ Res Public Health. 2010;7(8):3071-3079.

Agrawal

Platz

Niparko

JK.

Risk factors for hearing loss in US adults: data from the National Health and Nutrition Examination Survey, 1999 to 2002. Otol Neurotol. 2009;30(2):139-145.

Levine

Oron

Tinnitus. Handb Clin Neurol. 2015;129:409-431.

10.

Baguley

McFerran

Hall

Tinnitus. Lancet. 2013;382(9904):1600-1607.

11.

Langguth

Kreuzer

Kleinjung

De Ridder

Tinnitus: causes and clinical management. Lancet Neurol. 2013;12(9):920-930.

12.

Jarach

Lugo

Scala

, et al. Global prevalence and incidence of tinnitus: a systematic review and meta-analysis. JAMA Neurol. 2022;79(9):888-900.

13.

Jafari

Kolb

Mohajerani

MH.

Age-related hearing loss and tinnitus, dementia risk, and auditory amplification outcomes. Ageing Res Rev. 2019;56:100963.

14.

Yankaskas

Prelude: noise-induced tinnitus and hearing loss in the military. Hear Res. 2013;295:3-8.

15.

Kang

Kim

Yeo

SG.

Analysis of chronic tinnitus in noise-induced hearing loss and presbycusis. J Clin Med. 2021;10(8):1779.

16.

Turner

Larsen

Effects of noise exposure on development of tinnitus and hyperacusis: prevalence rates 12 months after exposure in middle-aged rats. Hear Res. 2016;334:30-36.

17.

Nondahl

Cruickshanks

Huang

, et al. Tinnitus and its risk factors in the Beaver Dam offspring study. Int J Audiol. 2011;50(5):313-320.

18.

Kvestad

Czajkowski

Engdahl

Hoffman

Tambs

Low heritability of tinnitus: results from the second Nord-Trondelag health study. Arch Otolaryngol Head Neck Surg. 2010;136(2):178-182.

19.

Khaldari

Khanjani

Bahrampour

Ghotbi Ravandi

Arabi Mianroodi

AA.

The relation between hearing loss and smoking among workers exposed to noise, using linear mixed models. Iran J Otorhinolaryngol. 2020;32(108):11-20.

20.

Rong

Wang

Lin

Association between smoking and noise-induced hearing loss: a meta-analysis of observational studies. Int J Environ Res Public Health. 2020;17(4):1201.

21.

Liu

Ding

, et al. Prevalence of hypertension and noise-induced hearing loss in Chinese coal miners. J Thorac Dis. 2016;8(3):422-429.

22.

Zhang

Ding

Yin

Zhang

Zhu

Research and discussion on the relationships between noise-induced hearing loss and ATP2B2 gene polymorphism. Int J Genomics. 2019;2019:5048943.

23.

Wang

Yang

Han

Huang

Sex differences in noise-induced hearing loss: a cross-sectional study in China. Biol Sex Differ. 2021;12(1):24.

24.

Jang

Kim

Kwon

HJ.

The association between impaired fasting glucose and noise-induced hearing loss. J Occup Health. 2011;53(4):274-279.

25.

Cantuaria

Pedersen

Poulsen

, et al. Transportation noise and risk of tinnitus: a nationwide cohort study from Denmark. Environ Health Perspect. 2023;131(2):27001.

26.

Akinbami

Chen

Davy

, et al. National Health and Nutrition Examination Survey, 2017-March 2020 prepandemic file: sample design, estimation, and analytic guidelines. Vital Health Stat 1. 2022;2022(190):1-36.

27.

Johnson

Paulose-Ram

Ogden

, et al. National health and nutrition examination survey: analytic guidelines, 1999-2010. Vital Health Stat 2. 2013;2013(161):1-24.

28.

Pan

Choi

Hokan

Doherty

JK.

Trends in hearing protection use with occupational noise exposure in the United States 1999 to 2016. Otol Neurotol. 2022;43(1):e14-e22.

29.

Carroll

Eichwald

Scinicariello

, et al. Vital signs: noise-induced hearing loss among adults—United States 2011-2012. MMWR Morb Mortal Wkly Rep. 2017;66(5):139-144.

30.

Kang

Cho

Park

, et al. Predictors for functionally significant in-stent restenosis: an integrated analysis using coronary angiography, IVUS, and myocardial perfusion imaging. JACC Cardiovasc Imaging. 2013;6(11):1183-1190.

31.

Schaette

Tinnitus in men, mice (as well as other rodents), and machines. Hear Res. 2014;311:63-71.

32.

Themann

Masterson

EA.

Occupational noise exposure: a review of its effects, epidemiology, and impact with recommendations for reducing its burden. J Acoust Soc Am. 2019;146(5):3879.

33.

Bhatt

Lin

Bhattacharyya

Prevalence, severity, exposures, and treatment patterns of tinnitus in the United States. JAMA Otolaryngol Head Neck Surg. 2016;142(10):959-965.

34.

Reisinger

Schmidt

Benz

, et al. Ageing as risk factor for tinnitus and its complex interplay with hearing loss-evidence from online and NHANES data. BMC Med. 2023;21(1):283.

35.

Oosterloo

Croll

Baatenburg

Jong

Ikram

Goedegebure

Prevalence of tinnitus in an aging population and its relation to age and hearing loss. Otolaryngol Head Neck Surg. 2021;164(4):859-868.

36.

Han

Lee

Suh

Lee

Park

MK.

Associations between tinnitus and body composition: a cross-sectional study. Sci Rep. 2024;14(1):16373.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB