Abstract
Objective
To examine the prevalence of cerumen impaction in a nationally representative sample of the US population and the association of cerumen impaction with sociodemographic factors, hearing loss, and tinnitus.
Methods
The cohort included 14,230 individuals aged ≥12 years who completed otoscopy and audiometry in NHANES (2005–2016). Cerumen impaction (partial/complete) was determined by otoscopy. Hearing level was defined by speech‐frequency pure‐tone average (PTA). Multivariable regression analysis was performed to identify factors associated with cerumen impaction.
Results
The prevalence of any cerumen impaction was 18.6% [95% CI: 17.3%–19.9%] among individuals ≥12 years and 32.4% [29.9%–35.1%] among those ≥70 years. The prevalence of bilateral partial and complete cerumen impaction was 6.3% [5.6%–7.1%] and 1.2% [1.0%–1.9%], respectively. Any cerumen impaction was associated with male sex (OR 1.77 [1.5–2.1]), identifying as Black race (vs. Caucasian, OR: 1.78 [1.5–2.9]), lower level of education (OR: 0.84 [0.71–0.98]), and older age (OR: 1.02 [1.01–1.03]). After adjusting for sociodemographic and clinical factors, complete impaction was associated with increased PTA (right ear: β = 4.1 dB [2.4–5.8 dB], left ear: β = 1.9 dB [0.46–3.4 dB]), but not with tinnitus.
Conclusions
Cerumen impaction is highly prevalent in the US population, especially among older adults, and has disproportionate sociodemographic impacts. Complete impaction is associated with a small, statistically significant elevation in PTA, but there is no association with tinnitus. These findings emphasize the need to implement and disseminate best practices for ear hygiene and cerumen management broadly and equitably.
Level of Evidence
2B
Keywords: CDC, cerumen, cerumen impaction, earwax, National Health and nutrition examination survey, NHANES, otology, prevalence
To examine the prevalence of cerumen impaction in a nationally representative sample of the US population and the association of cerumen impaction with sociodemographic factors, hearing loss and tinnitus. Overall cerumen impaction is highly prevalent in the US population, especially among older adults, and has disproportionate sociodemographic impacts. These findings emphasize the need to implement and disseminate best practices for ear hygiene and cerumen management broadly and equitably.
1. INTRODUCTION
Cerumen is a universal part of the human experience. It is a naturally occurring substance formed in the cartilaginous portion of the external auditory canal (EAC) that serves to lubricate, clean, and protect the ear. 1 Any impairment in the innate self‐cleaning mechanisms in the EAC can lead to excessive accumulation of cerumen, termed cerumen impaction, which is thought to cause aural fullness, pain, itching, and hearing loss. 2 , 3 The treatment of cerumen impaction is commonplace to many otolaryngology and primary care practices and represents a considerable healthcare burden. Annually, cerumen accounts for 12 million medical office visits 4 and cerumen removal is the third most billed procedure by otolaryngologists. 5 To guide evidence‐based practice, the AAO‐HNSF published a Clinical Practice Guideline (CPG) on cerumen impaction in 2008, with a subsequent update in 2017. 6 Reflecting the ongoing relevance of cerumen and cerumen impaction, the recommendations and resources provided in the CPG are the second most downloaded products from AAO‐HNSF, with over 130,000 views since 2016. 1
Despite evidence of significant healthcare utilization, few have explored the epidemiology of cerumen impaction and its key associations. A report based on a national sample of US adults from the National Health and Nutrition Examination Survey (NHANES) in 1999–2002 demonstrated the prevalence of cerumen impaction at 10%–12%. However, this report included only adults aged 20–69 years without adolescent or older adult populations. While prior studies have identified several factors associated with a higher burden of cerumen such as older age and individuals with intellectual disabilities, there is currently limited knowledge regarding other demographic or clinical factors that may play a role. 7 , 8 , 9 Similarly, despite hearing loss and tinnitus being frequently cited symptoms of cerumen impaction, the supporting data is not as established. Investigations into the audiometric impact of cerumen have demonstrated an elevation in pure tone thresholds with increasing levels of EAC occlusion, 7 but the extent of this effect remains disputed as identical levels of occlusion have resulted in anywhere from a 5 to 40 dB hearing loss. 10 , 11 The relationship between tinnitus and cerumen is even more ambiguous, with no primary data to support an association. 12 , 13 Greater clarification of these relationships can guide the clinical work‐up of patients who present with these symptoms.
In this study, we aimed to estimate the prevalence of cerumen impaction and identify associated demographic and clinical factors in a national sample of US population aged 12 years and older. Additionally, to inform clinical practice and counseling, we aimed to quantify the association of cerumen impaction with tinnitus and audiometrically measured hearing loss.
2. METHODS
2.1. Study population
The study cohort was comprised of 14,230 participants in the NHANES aged 12 years and older. NHANES is a comprehensive collection of questionnaires and physical examinations designed to investigate the health status of US adults and children. Data were combined from five cycles, years 2005 through 2016 (excluding cycle 2013–2014 with no available audiometry data). Study of the de‐identified publicly available dataset was deemed exempt by the University of Minnesota Institutional Review Board (#00015174).
2.2. Primary outcome: Cerumen impaction
Prior to audiometric assessment, trained NHANES examiners performed bilateral otoscopic examination to assess for physical abnormalities of each ear canal (e.g., collapsing canal, cerumen) or tympanic membrane (e.g., perforation). 14 Excessive cerumen, hereby referred to as partial impaction, was defined as significant accumulation of cerumen such that less than half of the tympanic membrane can be seen as it was partially (but not completely) blocked by cerumen. Complete cerumen impaction was defined as accumulation of cerumen blocking any visualization of the tympanic membrane. These definitions are supported by the NHANES 2018 Policy and Procedures manual which gives photo examples of each definition to guide reviewers. 14 The main outcome “any cerumen impaction” was defined as having partial or complete cerumen impaction in either ear. The reported prevalence values refer to point prevalence or the proportion at one point in time.
2.3. Covariates
As part of an otologic symptom questionnaire, the presence of tinnitus was assessed with the question “in the past 12 months have you been bothered by ringing, roaring, or buzzing in your ears or head that lasts for five minutes or more?” Air conduction audiometry assessments were conducted based on established NHANES protocols. 14 Ear‐specific (left and right) speech frequency pure tone average (PTA) was calculated from the thresholds at 500, 1000, 2000, and 4000 Hz.
Covariates of interest were selected based on relevance to social determinants of health and otologic health. Covariates included sex (binary); race/ethnicity (grouped as non‐Hispanic white, non‐Hispanic Black, Hispanic (Mexican American or other Hispanic), or Asian); education level (<12th grade, high school graduate, some college or more); household income (<$20,000; $20,000–$44,999; $45,000–$74,999; ≥$75,000/unknown); and insurance (none; private health plan; Medicare; Medicaid; unknown). Medical history variables included diabetes (DM) (self‐reported diagnosis and/or current use of insulin or other diabetic medications), hypertension (HTN) (self‐reported diagnosis and/or current use of anti‐hypertensive medications), cardiovascular disease (congestive heart failure, coronary artery disease, angina pectoris, or myocardial infarction), stroke (self‐reported), smoking status (current, former, and never), and hearing aid use (self‐reported). Body mass index (BMI) was categorized into the five World Health Organization (WHO) weight classes (underweight BMI <18.5 kg/m2; normal weight 18.5–24.9 kg/m2; overweight 25–29.9 kg/m2; obesity class I 30–34.9 kg/m2; obesity class II 35–39.9 kg/m2; obesity class III >40 kg/m2). 14 , 15 History of occupational noise exposure, recreational noise exposure, and hearing protection use were included in multivariable regression models for outcomes assessing hearing loss and tinnitus. These variables were defined as binary variables as in prior study. 16 , 17
A history of hearing protection use was missing in a subset of the cohort (n = 1890, 13.3%). Additional analysis was performed to investigate sociodemographic factors associated with the missingness of hearing protection use. This analysis found hearing protection use data were more likely to be missing among younger, female (vs. male), Asian (vs. White) participants, and participants with higher education and private insurance (vs. Medicare).
Sensitivity analysis showed that there were no significant changes in results or effect size with and without hearing protection use as a covariate.
2.4. Statistical analysis
The complete dataset underwent sample weighting according to the established National Center for Health Statistics guidelines to yield results generalizable to the US population. Unweighted data are exclusively reported in Table 1 providing descriptive statistics for the study cohort. Pearson χ2 test and Student t‐test were used to compare baseline characteristics for categorical and continuous variables, respectively. Univariable and multivariable logistic regression analyses were used to identify factors associated with cerumen impaction. Multivariable regression models adjusted for age, sex, race/ethnicity, education, income, insurance, smoking status, BMI, DM, and HTN. In multivariable regression analysis assessing the association of cerumen impaction with hearing loss and tinnitus, occupational/recreational noise exposure, and hearing protection were added as covariates when they were significantly associated with the outcomes of hearing loss and tinnitus. Collinearity diagnostics were performed for all explanatory variables, using a cutoff variance inflation factor (VIF); a measure of the amount of multicollinearity in regression analysis of >2.5 as problematic correlation among variables a priori. A history of stroke and cardiovascular disease were excluded from the analysis due to their VIF values >2.5. Subsequent collinearity diagnostics performed after exclusion of stroke and cardiovascular disease history showed no evidence of collinearity among covariates. Linear regression analysis was used to examine the association between cerumen impaction and PTA, separated by ears. Logistic regression analysis was used to assess the association between various levels of cerumen impaction (unilateral partial, bilateral partial, unilateral complete, and bilateral complete impaction) and tinnitus. Statistical significance was set at p < .05, two‐tailed. Analyses were conducted with Stata (version 16.1; StataCorp, College Station, TX).
TABLE 1.
Unweighted sample characteristics by Cerumen impaction in US adults (N = 14,230).
| Any cerumen impaction | ||||
|---|---|---|---|---|
| Overall N = 14,230 | Yes | No | p‐value | |
| Age, year, average (SD) | 39 (21.8) | 43.7 (23.8) | 37.8 (21.0) | <.001 |
| Sex, male, n (%) | 7150 (50.3) | 1729 (57.3) | 5421 (48.4) | |
| Race, n (%) | <.001 | |||
| White | 5104 (35.9) | 1050 (34.8) | 4054 (36.2) | |
| Black | 3510 (24.7) | 945 (31.3) | 2565 (22.9) | |
| Hispanic | 3981 (28.0) | 753 (24.9) | 3228 (28.8) | |
| Asian | 1074 (7.5) | 166 (5.5) | 908 (8.1) | |
| Other | 561 (3.9) | 106 (3.5) | 455 (4.0) | |
| Education, n (%) | <.001 | |||
| Less than high school | 2275 (16.0) | 585 (19.4) | 1690 (15.1) | |
| High school graduate | 2178 (15.3) | 542 (17.9) | 1636 (14.6) | |
| Some college or more | 5, 372 (37.7) | 1056 (35.0) | 4, 316 (38.5) | |
| Refused/do not know | 4405 (31.0) | 837 (27.7) | 3, 568 (31.8) | |
| Income, n (%) | <.001 | |||
| <20k | 3043 (21.4) | 722 (23.9) | 2321 (20.7) | |
| 20–44.9k | 4036 (28.3) | 915 (30.3) | 3121 (27.8) | |
| 45–74.9k | 2645 (18.6) | 550 (18.2) | 2095 (18.7) | |
| >75k | 3400 (23.9) | 612 (20.3) | 2788 (24.9) | |
| Refused/do not know/unsure | 1106 (7.8) | 221 (7.3) | 885 (7.9) | |
| Insurance, n (%) | <.001 | |||
| None | 2764 (19.4) | 528 (17.5) | 2236 (19.9) | |
| Private | 6051 (42.5) | 1089 (36.0) | 4962 (44.3) | |
| Medicare | 2062 (14.5) | 655 (21.7) | 1407 (12.6) | |
| Medicaid | 1850 (13.0) | 461 (15.3) | 1389 (12.4) | |
| Unknown | 1503 (10.6) | 287 (9.5) | 1216 (10.8) | |
| Smoking, n (%) | <.001 | |||
| Never | 5557 (39.1) | 1157 (38.3) | 4400 (39.3) | |
| Former | 2377 (16.7) | 581 (19.2) | 1796 (16.0) | |
| Current | 1884 (13.2) | 445 (14.8) | 1439 (12.8) | |
| Unknown | 4412 (31.0) | 837 (27.7) | 3575 (31.9) | |
| BMI, n (avg; SD) | .405 | |||
| Normal | 5021 (35.3) | 1, 045 (34.6) | 3, 976 (35.5) | |
| Underweight | 726 (5.1) | 149 (4.9) | 577 (5.15) | |
| Overweight | 3959 (27.8) | 836 (27.7) | 3123 (27.9) | |
| Obesity class I | 2438 (17.1) | 536 (17.8) | 1902 (17.0) | |
| Obesity class II | 1117 (7.9) | 247 (8.2) | 870 (7.8) | |
| Obesity class III | 841 (5.9) | 171 (5.7) | 670 (6.0) | |
| Diabetes, n (%) | <.001 | |||
| Non‐diabetic | 12,843 (90.3) | 2633 (87.2) | 10,210 (91.1) | |
| Diabetic | 1387 (9.7) | 387 (12.8) | 1000 (8.9) | |
| Hypertension, n (%) | <.001 | |||
| No | 8602 (60.5) | 1704 (56.4) | 6898 (61.5) | |
| Yes | 3417 (24.0) | 882 (29.2) | 2535 (22.6) | |
| Unknown | 2211 (15.5) | 434 (14.4) | 1777 (15.9) | |
| Cardiovascular disease, n (%) | <.001 | |||
| No | 9012 (63.3) | 1896 (62.8) | 7116 (63.5) | |
| Yes | 814 (5.7) | 287 (9.5) | 527 (4.7) | |
| Unknown | 4404 (31.0) | 837 (27.7) | 3567 (31.8) | |
| Stroke, n (%) | <.001 | |||
| No | 9449 (66.4) | 2062 (68.3) | 7387 (65.9) | |
| Yes | 369 (2.6) | 119 (3.9) | 250 (2.2) | |
| Unknown | 4412 (31.0) | 839 (27.8) | 3573 (31.9) | |
| Hearing aid use, n (%) | .081 | |||
| No | 14,111 (99.2) | 2987 (98.9) | 11,124 (99.2) | |
| Yes | 119 (0.8) | 33 (1.1) | 86 (0.8) | |
Note: Any cerumen impaction includes participants classified as excessive impaction or complete impaction.
Abbreviations: BMI, body mass index; SD, standard deviation.
3. RESULTS
Unweighted cohort characteristics are presented in Table 1. After accounting for sampling weights, 15 the overall US prevalence of any cerumen impaction was estimated to be 18.6% (95% CI: 17.4–19.9) for those ≥12 years and 32.4% (29.9%–35.1%) for older adults (≥ 70 years). Figure 1. summarizes the prevalence of partial and complete impaction for individual and bilateral ears. In all analyzed subgroups, the prevalence of partial cerumen impaction was higher than that of complete cerumen impaction for both unilateral and bilateral states. There was no significant difference between right and left ears for either level of impaction.
FIGURE 1.
(A) and (B) Prevalence of cerumen impaction in the United States. Any cerumen impaction includes diagnosis of both partial and complete impaction. (°cerumen obstructing >50% of view of tympanic membrane on otoscopy. *cerumen completely blocking view of tympanic membrane on otoscopy.)
Factors associated with cerumen impaction are presented in Table 2. In univariable analyses, any cerumen impaction was significantly associated with older age, male sex, Black race, lower education, lower income, having Medicare/Medicaid insurance, and history of smoking, DM, and HTN. In the multivariable model, factors independently associated with any cerumen impaction were age, sex, race, education, and insurance status. Older age was associated with higher odds of cerumen impaction, with a 20% increased likelihood of impaction per decade of age (OR: 1.02 [1.01–1.03]). Male participants were 1.8 times more likely to have cerumen impaction than female participants (OR 1.77 [1.53–2.05]). Black participants were 1.8 times more likely to have cerumen impaction than white participants (OR: 1.78 [1.53–2.89]). Participants with less than a high school education were 1.2 times more likely to have cerumen impaction than participants with some college education or more (OR 1.16 [1.02–1.29]). The participants enrolled in Medicare and Medicaid insurance were 1.4 times (OR: 1.42 [1.11–1.82]) and 1.3 times (OR: 1.34 [1.09–1.64]) more likely to have cerumen impaction than counterparts with private insurance. Medical factors, including DM and HTN, were not associated with cerumen impaction in the multivariable model. Hearing aid use was not associated with cerumen impaction in univariable or multivariable models.
TABLE 2.
Univariable and multivariable logistic regression analyses examining factors associated with any cerumen impaction.
| Univariable | Multivariable | |||
|---|---|---|---|---|
| Odds ratio (95% CI) | p‐value | Odds ratio (95% CI) a | p‐value | |
| Age | 1.02 (1.01–1.02)** | <.001 | 1.02 (1.01–1.02) | <.001 |
| Sex b | 0.60 (0.53–0.69)** | <.001 | 0.57 (0.49–0.65) | <.001 |
| Race | ||||
| White | Ref | Ref | ||
| Black | 1.65 (1.43–1.89)** | <.001 | 1.79 (1.53–2.09)** | <.001 |
| Hispanic | 1.01 (0.88–1.1.6) | .835 | 1.15 (0.98–1.34) | .085 |
| Asian | 0.81 (0.66–0.98)* | .034 | 0.94 (0.76–1.17) | .590 |
| Others | 0.99 (0.76–1.30) | .938 | 1.04 (0.77–1.41) | .786 |
| Education | ||||
| Less than high school | Ref | Ref | ||
| High school graduate | 0.87 (0.69–1.09) | .222 | 0.97 (0.76–1.24) | .797 |
| Some college or more | 0.65 (0.56–0.75)** | <.001 | 0.84 (0.72–0.98)* | .032 |
| Refused/do not know | 0.62 (0.52–0.74)** | <.001 | 0.53 (0.12–2.41) | .408 |
| Income | ||||
| <20k | Ref | Ref | ||
| 20–44.9k | 0.94 (0.78–1.14) | .551 | 1.05 (0.85–1.31) | .628 |
| 45–74.9k | 0.84 (0.66–1.05) | .129 | 1.02 (0.78–1.33) | .892 |
| >75k | 0.70 (0.59–0.84)** | <.001 | 0.97 (0.77–1.21) | .774 |
| Refused/do not know/unsure | 0.87 (0.67–1.12) | .270 | 0.97 (0.76–1.24) | .819 |
| Insurance | ||||
| Private | Ref | Ref | ||
| None | 0.90 (0.76–1.07) | .239 | 1.01 (0.84–1.22) | .927 |
| Medicare | 2.08 (1.73–2.50)** | <.001 | 1.40 (1.10–1.78)** | .007 |
| Medicaid | 1.42 (1.19–1.70)** | <.001 | 1.33 (1.08–1.63)** | .008 |
| Unknown | 0.98 (0.79–1.20) | .811 | 0.99 (0.81–1.20) | .889 |
| Smoking | ||||
| Never | Ref | 0.94 (0.79–1.13) | .528 | |
| Current | 1.21 (0.99–1.47) | .066 | 0.31 (0.04–2.61) | .277 |
| Former | 1.24 (1.07–1.45)** | .006 | 1.06 (0.88–1.29) | .541 |
| BMI | ||||
| Normal | Ref | 1.22 (0.89–1.68) | .215 | |
| Underweight | 1.10 (0.83–1.48) | .471 | 0.88 (0.72–1.09) | .238 |
| Overweight | 1.03 (0.85–1.26) | .726 | 0.88 (0.74–1.04) | .142 |
| Obesity class I | 1.05 (0.90–1.24) | .528 | 0.90 (0.70–1.14) | .366 |
| Obesity class II | 1.02 (0.81–1.29) | .863 | 0.92 (0.69–1.23) | .567 |
| Obesity class III | 1.01 (0.77–1.32) | .961 | 1.28 (0.74–2.21) | .375 |
| Diabetes | 1.54 (1.31–1.81)** | <.001 | 1.07 (0.88–1.31) | .480 |
| Hypertension | 1.15 (1.06–1.24)** | <.001 | 0.97 (0.80–1.18) | .771 |
| Cardiovascular disease | 0.98 (0.91–1.05) | .533 | 1.16 (0.93–1.31) | .189 |
| Stroke | 0.94 (0.87–1.01) | .095 | 1.03 (0.81–1.31) | .789 |
| Hearing aid use | 1.41 (0.78–2.54) | .257 | 1.06 (0.59–1.92) | .844 |
Note: Crude unadjusted model.
Abbreviations: CI, confidence interval; Ref, reference.
Multivariable regression analysis was adjusted for associated variables (age, sex, race, education, income, insurance, smoking status, BMI, diabetes mellitus, hypertension, cardiovascular disease, stroke history).
Male as a reference group.
p < .05;
p < .01.
In multivariable analysis, individuals with complete impaction were found to have 4.10 dB (2.37–5.83) higher PTA in the right ear and 1.93 dB (0.46–3.38) higher PTA in the left ear compared to individuals with no impaction (Table 3). When analyzed against the PTA in no impaction states, partial cerumen impaction did not significantly affect PTA (Figure 2.). Clinical significance of the elevated PTA in this range is unclear. In multivariable analyses to assess the relationship between cerumen impaction and tinnitus (Table 4), neither unilateral nor bilateral partial or complete cerumen impaction were associated with the presence of tinnitus.
TABLE 3.
| Right | Left | |||
|---|---|---|---|---|
| Cerumen impaction | Beta‐coefficient (95% CI) | p‐value | Beta‐coefficient (95% CI) | p‐value |
| Partial impaction | 0.94 (−0.08 to 1.96) | .071 | 0.21 (−0.72 to 1.15) | .653 |
| Complete impaction | 4.34 (2.65–6.03)** | <.001 | 2.73 (0.66–4.79)* | .010 |
PTA = pure tone average, calculated from thresholds at 500, 1000, 2000, and 4000 Hz.
Model adjusted by all relevant variables (age, gender, race, education, income, insurance, smoking status, BMI, diabetes mellitus, hypertension, cardiovascular disease, stroke history).
p < .05;
p < .01.
FIGURE 2.
(A) and (B). Box plot of right and left ear PTA at different levels of impaction.
TABLE 4.
Multivariable logistic regression analysis assessing the association between cerumen impaction and tinnitus. a
| Partial Cerumen impaction | Complete Cerumen impaction | |||
|---|---|---|---|---|
| Odds ratio (95% CI) | p‐value | Odds ratio (95% CI) | p‐value | |
| No impaction | Ref | Ref | ||
| Unilateral | 0.98 [0.72–1.34] | .895 | 0.64 [0.39–1.07] | .087 |
| Bilateral | 0.91 [0.69–1.20] | .505 | 0.92 [0.54–1.56] | .747 |
Model adjusted by all relevant variables (PTA from better hearing ear, age, gender, race, education, income, insurance, smoking status, BMI, diabetes mellitus, hypertension, cardiovascular disease, stroke history).
4. DISCUSSION
In a nationally representative sample of the US population, we observed an 18.6% prevalence of any cerumen impaction among individuals ≥12 years and 32.4% prevalence among adults ≥70 years. Partial impaction was approximately five times more common than complete impaction, in both unilateral and bilateral conditions, with no significant difference between ears. Cerumen impaction was significantly associated with older age, male sex, Black race, public insurance, and lower education, suggesting a link between cerumen impaction and social determinants of health.
Our study demonstrated high prevalence of cerumen impaction in the United States, especially among older adults. The estimated prevalence was greater than reports from prior studies based on communities of specific regions in the United States ranging from 5% to 8%. 7 , 12 One prior study based on analysis of NHANES cohorts from 1999 to 2002 reported a 10%–12% prevalence. The prevalence difference between the study of prior NHANES cohorts and the current study is likely due to the addition of older adults in our cohort who were found to have significantly higher likelihood of any cerumen impaction. The prevalence of cerumen impaction in our cohort limiting age to 20–69 years as in the prior study showed no substantial changes in prevalence in this age group over time. Older adults tend to have higher prevalence of cerumen impaction due to multiple factors such as changes in ear canal shape, increased hair growth, and decreased self‐cleaning ability of the ear. Addressing cerumen impaction, especially among older adults has clinical implications for optimizing their hearing thresholds and reducing associated symptoms. 6 , 8 Routine monitoring and appropriate cerumen cleaning should be offered when necessary.
We observed several sociodemographic patterns regarding cerumen impaction. Within our dataset, males were twice as likely to have cerumen impaction as compared to females. This is consistent with the findings of community‐based studies and aligns with the theory that the larger, coarser hair found at the external auditory meatus in men impairs the egress of wax. 18 , 19 Similar changes to the meatal hair also occur with aging as above. 3 Our findings reinforce the downstream effects of these physiologic changes, as each decade of age corresponded with a 20% increased likelihood of cerumen impaction. Analysis of additional demographic factors revealed significant differences in cerumen impaction by race, education, and insurance status. The increased likelihood of cerumen impaction observed in Black participants is likely multifactorial. The genetic origin of cerumen phenotype is well‐established, with the wet (vs. dry) phenotype more frequently observed in those of African ancestry. 20 Presumably, the force required for canal skin to extrude wet cerumen with its characteristic “sticky” quality is greater than the force required for the “flaky” dry wax, 21 increasing risk of cerumen impaction. There may also be different culturally influenced ear‐cleaning practices. 22
Patterns of healthcare engagement may underpin some of our observations. Compared to Caucasians, Black Americans have lower rates of ambulatory care usage and lower overall healthcare expenditure. 23 A similar pattern is seen in Americans of lower socioeconomic status (SES), as individuals of low SES use substantially less ambulatory and preventative care than high‐SES patients. 22 , 23 , 24 These patterns of healthcare utilization may partially account for the increased likelihood of cerumen impaction in participants with lesser education or those enrolled in public insurance (Medicare or Medicaid), both independent markers of lower SES. As seen in other medical conditions, reduced engagement in ambulatory services limits health literacy, preventative care, and treatment of precursor conditions (e.g., excessive cerumen). 22 , 23 , 24 The identification of higher‐risk populations can guide future education efforts and ultimately improve health equity and access to care for cerumen impaction. Fortunately, there are over‐the‐counter and low‐cost treatment options for cerumen that offer safe and realistic tools for at‐home management, eliminating many of the barriers to care faced with other health conditions. There should be efforts to increase the general knowledge of and recommendations regarding such options. In conjunction with a CPG, the AAO‐HNS has published plain language guidelines for cerumen impaction which could be purposefully disseminated among populations and regions with a high density of at‐risk persons. 1 Interestingly, there was no significant association between hearing aid use and cerumen impaction in this cohort. Hearing aid use itself may increase the likelihood of cerumen dislocation into the medial canal by causing canal blockage. On the other hand, hearing aid users are likely routinely seen by hearing care professionals and may have their cerumen debrided more often. Given the limited data in this study on hearing aid use and debridement pattern, we are unable to make a conclusion on the impact of hearing aid use on cerumen impaction.
Our findings suggest that complete cerumen impaction can increase hearing thresholds. The clinical significance of increased hearing thresholds in range of 1.93–4.10 dB is unclear. Prior studies have observed sound intensity difference as low as 4 dB may be noticeable to individuals with normal hearing. 25 , 26 Partial impaction had no significant impact on hearing thresholds, aligning with prior studies in which hearing changes occurred when cerumen involved at least 80% of the cross‐sectional area of the ear canal. 27 Contrary to what is frequently reported by patients and in non‐primary sourced literature, after adjusting for relevant factors there was no association between the presence of tinnitus and cerumen impaction. This finding emphasizes the importance of clinical investigation of the cause of tinnitus, particularly when it is bothersome or asymmetric, rather than concluding that observed cerumen is the source. 28
There are limitations to our study. As a cross‐sectional study, we were unable to examine temporal relationships between cerumen impaction and variables of interest. The definition of cerumen impaction used in prior literature varies, thus limiting precise comparisons. In attempts to overcome this, cerumen impaction was exclusively based on the degree of tympanic membrane view obstruction. Cerumen impaction was assessed using otoscopy by non‐clinicians and microscopic exams were unavailable. The detailed examination of cerumen (e.g., depth, location) was not possible in this study. While standardized protocols were employed as described in the procedure protocol ratings of cerumen impaction as partial or complete may be dependent on the rater. 14 The presence or absence of symptoms related to excessive cerumen was not considered in our diagnostic criteria (e.g., otalgia, aural fullness, decreased hearing). Thus, our prevalence data directly reflects the number of Americans afflicted with cerumen impaction. From our clinical experience, we suspect that a significant number of individuals seek medical care for such symptoms but have otoscopy findings that would not qualify as cerumen impaction (>50% obstruction of TM view). As such, our 18.6% prevalence of cerumen impaction likely underestimates the percentage of Americans who may utilize healthcare resources for cerumen disimpaction.
As a result of sourcing our data from a national database we can control for some but not all confounding variables. Objective information regarding the presence of ear canal pathologies (e.g., canal size, canal osteoma, and exostoses) that may contribute to cerumen impaction is not available. Additionally, given the paucity of published literature investigating environmental and physiologic factors related to cerumen burden, our analysis may have excluded confounding variables that are currently unknown. We also could not control for all variables associated with hearing loss or tinnitus due to limited availability (e.g., exposure to ototoxic medication or chemicals, surgical‐associated hearing loss, and medical history causing hearing loss). We were able to incorporate hearing protection use and loud noise exposure (both occupational and recreational) into this analysis. The missing data on hearing protection use in a subset of our cohort was associated with sociodemographic factors and may represent a potential source of bias. We could not account for individual hygiene practices or in‐ear device usage other than hearing aids, and bone conduction thresholds were not available to differentiate conductive from sensorineural hearing loss.
5. CONCLUSION
Cerumen impaction is highly prevalent in the US population, especially among older adults. Cerumen impaction is associated with hearing loss and disproportionately impacts individuals based on social determinants of health and SES, particularly minority race, lower education, and public insurance. Improving education on best practices for aural hygiene and cerumen management for higher‐risk populations may increase engagement with safe and effective tools for at‐home treatment. We also must improve access to the resources provided in the AAO‐HNS CPGs and ensure they are utilized in ambulatory and emergency care practices.
FUNDING INFORMATION
Dr. Tolan reported receiving grants from the Lions Hearing Foundation for the submitted work.
CONFLICT OF INTEREST STATEMENT
Dr. Adams served on a medical advisory council to Advanced Bionics, independent of this research work.
Tolan MM, Choi JS, Tibesar MT, Adams ME. Cerumen impaction: Prevalence and associated factors in the United States population. Laryngoscope Investigative Otolaryngology. 2024;9(2):e1228. doi: 10.1002/lio2.1228
REFERENCES
- 1. Roland PS, Smith TL, Schwartz SR, et al. Clinical practice guideline: Cerumen impaction. Otolaryngol‐Head Neck Surg. 2008;139(3):S1‐S21. [DOI] [PubMed] [Google Scholar]
- 2. Subha ST, Raman R. Role of impacted cerumen in hearing loss. Ear Nose Throat J. 2006;85(10):650‐653. [PubMed] [Google Scholar]
- 3. Roeser RJ, Ballachanda BB. Physiology, pathophysiology, and anthropology/epidemiology of human ear canal secretions. J Am Acad Audiol. 1997;8:391‐400. [PubMed] [Google Scholar]
- 4. Mitka M. Cerumen removal guidelines wax practical. JAMA Otolaryngol Head Neck Surg. 2008;300(13):1506. [DOI] [PubMed] [Google Scholar]
- 5. Yang EL, Macy TM, Wang KH, Durr ML. Economic and demographic characteristics of cerumen extraction claims to Medicare. JAMA Otolaryngol Head Neck Surg. 2016;142(2):157‐161. [DOI] [PubMed] [Google Scholar]
- 6. Schwartz SR, Magit AE, Rosenfeld RM, et al. Clinical Practice Guideline (Update): Earwax (Cerumen Impaction). Otolaryngol Head Neck Surg. 2017;156(1_suppl):S1‐S29. doi: 10.1177/0194599816671491 Erratum in: Otolaryngol Head Neck Surg. 2017 Sep; 157(3):539. [DOI] [PubMed] [Google Scholar]
- 7. Roeser RJ, Lai L, Clark JL. Effect of ear canal occlusion on pure‐tone threshold sensitivity. J Am Acad Audiol. 2005;16(9):740‐746. [DOI] [PubMed] [Google Scholar]
- 8. Meyer F, Preuß R, Angelow A, Chenot JF, Meyer E, Kiel S. Cerumen impaction removal in general practices: a comparison of approved standard products. J Prim Care Community Health. 2020;11:2150132720973829. doi: 10.1177/2150132720973829 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Sugiura S, Yasue M, Sakurai T, et al. Effect of cerumen impaction on hearing and cognitive functions in Japanese older adults with cognitive impairment. Geriatr Gerontol Int. 2014;14(S2):56‐61. [DOI] [PubMed] [Google Scholar]
- 10. Willems M, van Berlaer G, Maes L, Leyssens L, Koehler B, Marks L. Outcome of 10 years of ear and hearing screening in people with intellectual disability in Europe: a multicentre study. J Appl Res Intellect Disabil. 2022;35(1):123‐133. [DOI] [PubMed] [Google Scholar]
- 11. Narayanan DA, Raman R, Chong AW. The role of occlusion of the external ear canal in hearing loss. Turk Arch Otorhinolaryngol. 2019;57(3):122‐126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. McManus B, Harbarger C, Grillis A, et al. Otoscopy and tympanometry outcomes from the National Health and nutrition examination survey (NHANES). Am J Otolaryngol. 2022;43(2):103332. [DOI] [PubMed] [Google Scholar]
- 13. Sharp JF, Wilson JA, Ross L, Barr‐Hamilton RM. Earwax removal: a survey of current practice. BMJ. 1990;301:1251‐1253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Centers for Disease Control and Prevention . National Health and Nutrition Examination Survey Data and Examination Protocol (2002‐2019). US Department of Health and Human Services, Centers for Disease Control and Prevention; 2019. https://wwwn.cdc.gov/nchs/nhanes/continuousnhanes/manuals.aspx?BeginYear=2019 [Google Scholar]
- 15. Weir CB, Jan A. BMI Classification Percentile and Cut off Points. StatPearls Publishing; 2022. [PubMed] [Google Scholar]
- 16. Pan DW, Choi JS, Hokan A, 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. doi: 10.1097/MAO.0000000000003343 [DOI] [PubMed] [Google Scholar]
- 17. Midi H, Sarkar SK, Rana S. Collinearity diagnostics of binary logistic regression model. J Interdiscip Math. 2010;13(3):253‐267. [Google Scholar]
- 18. Hanger HC, Mulley GP. Cerumen: its fascination and clinical importance: a review. J Roy Soc Med. 1992;85:346‐349. [PMC free article] [PubMed] [Google Scholar]
- 19. Guest JF, Greener MJ, Robinson AC, Smith AF. Impacted cerumen composition, production, epidemiology and management. QJM. 2004;97(8):477‐488. [DOI] [PubMed] [Google Scholar]
- 20. Dickman SL, Gaffney A, McGregor A, et al. Trends in health care use among black and white persons in the US, 1963‐2019. JAMA Netw Open. 2022;5(6):e2217383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Adegbiji WA, Alabi BS, Olajuyin OA, Nwawolo CC. Earwax impaction: symptoms, predisposing factors and perception among Nigerians. J Fam Med Primary Care. 2014;3:379‐382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Kleinman JC, Gold M, Makuc D. Use of ambulatory medical care by the poor: another look at equity. Med Care. 1981;19(10):1011‐1029. [DOI] [PubMed] [Google Scholar]
- 23. Aday LA, Andersen RM. The national profile of access to medical care: where do we stand? Am J Public Health. 1984;74(12):1331‐1339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Bloom B. Use of selected preventive care procedures in the United States. Vital Health Stat 10. 1986;74(157):1‐43. [PubMed] [Google Scholar]
- 25. Abbott D. Decibels and sound levels. Understanding Sound. Pressbooks; 2019:41. [Google Scholar]
- 26. Jesteadt W, Wier CC, Green DM. Intensity discrimination as a function of frequency and sensation level. J Acoust Soc Am. 1977. Jan;61(1):169‐177. doi: 10.1121/1.381278 [DOI] [PubMed] [Google Scholar]
- 27. Chandler JR. Partial occlusion of the external auditory meatus: its effect upon air and bone conduction hearing acuity. Laryngoscope. 1926;74:22‐53. [DOI] [PubMed] [Google Scholar]
- 28. Tunkel DE, Bauer CA, Sun GH, Rosenfeld RM. Clinical practice guideline: tinnitus. Otolaryngol Head Neck Surg. 2014;151(2 Suppl):S1‐S40. [DOI] [PubMed] [Google Scholar]