Comorbidity Patterns Between Hearing Loss and Symptomatic Dizziness in Middle‐Aged and Older Adults

ABSTRACT

Objective

To explore the comorbidity patterns between hearing loss and symptomatic dizziness (HL‐SD) in middle‐aged and older adults.

Methods

This cross‐sectional study used data from the 1999 to 2004 National Health and Nutrition Examination Survey (NHANES). After excluding incomplete data, 2961 participants aged 40–69 were analyzed. Logistic regression models, restricted cubic splines (RCS), and subgroup analyses assessed the association between hearing thresholds and dizziness. Univariate and multivariate logistic models and LASSO regression identified risk factors related to HL‐SD. Propensity score matching (PSM) evaluated the correlation between hearing loss and dizziness treatment efficacy.

Results

Among participants, 28.67% had hearing loss, and 22.99% reported dizziness, with 8.00% experiencing both. Increased hearing thresholds were significantly associated with a higher risk of dizziness, with each 1 dB increase in speech‐frequency PTA increasing dizziness risk by 1.6%. Various factors, including gender, income, arthritis, and diabetes, were linked to the comorbidity of HL‐SD. However, hearing loss did not significantly affect dizziness treatment outcomes.

Conclusions

Hearing loss is significantly associated with an increased risk of symptomatic dizziness in middle‐aged and older adults, influenced by various factors. However, it does not appear to affect dizziness treatment efficacy.

Summary

• Question: Is the association between hearing and symptomatic dizziness among individuals aged 40–69 years, what are the risk factors for comorbidity, and whether hearing loss affects the treatment effect of symptomatic dizziness?

• Findings: In this cross‐sectional study of 2961 US individuals, hearing loss was associated with symptomatic dizziness. Females, living with partner, BMI, neck pain, joint pain, hypertension, diabetes, headaches and tinnitus associated with the comorbidity. There was no significant correlation between hearing loss and dizziness treatment effect.

• Meaning: The findings quantified the association between hearing and symptomatic dizziness, and provided new insights for the prevention of comorbidities.

1. Introduction

Hearing loss is a significant global public health issue. Currently, about one‐fifth of the global population suffers from hearing loss [1]. According to the World Health Organization (WHO), the number of people with disabling hearing loss is expected to reach 900 million by 2050 [2]. Hearing loss is often linked to various diseases [3, 4, 5], particularly those frequently accompanied by dizziness. Dizziness has a lifetime prevalence of 15% to 36% in the population [6, 7, 8, 9], making it one of the most common clinical symptoms. Concurrent hearing loss and symptomatic dizziness significantly impact quality of life and pose a serious public health burden [10, 11].

Since the 19th century, researchers have identified a link between hearing loss and symptomatic dizziness (HL‐SD) [12]. Over the past century, it has been observed that hearing loss associated with specific inner ear diseases, such as Ménière's Disease (MD) and Sudden Sensorineural Hearing Loss, often occurs alongside or sequentially with symptomatic dizziness [13, 14, 15, 16, 17, 18, 19]. However, comprehensive quantitative analysis of the comorbid association across all types and degrees of hearing loss and symptomatic dizziness is lacking. Thus, a detailed epidemiological study is needed to clarify this issue.

In this cross‐sectional study, we assessed the prevalence of HL‐SD among middle‐aged and older adults in the United States. We quantified the association between increasing auditory threshold levels and the risk of symptomatic dizziness, identified comorbid risk factors, and investigated the impact of hearing loss on the prognosis of symptomatic dizziness. This enhanced our understanding of the comorbidity pattern between HL‐SD.

2. Methods

2.1. Study Population

The National Health and Nutrition Examination Survey (NHANES), sponsored by the National Center for Health Statistics (NCHS), evaluates the health and nutritional status of the American population. It also assesses related health behaviors and risk factors. NHANES collects comprehensive data through face‐to‐face interviews and physical examinations, covering health conditions, medical history, nutritional intake, body measurements, and biochemical metrics. All participants provided informed written consent to safeguard their rights.

To investigate the association between HL‐SD, we utilized NHANES data from 1999 to 2004. This data included symptomatic dizziness questionnaire responses and hearing data from middle‐aged and older participants. Out of 31,126 participants surveyed, 21,156 were excluded for being under 40 years old, and 3267 were excluded for being over 69 years old. Of the remaining 6694 participants, one was excluded due to missing symptomatic dizziness questionnaire data, and 3732 were excluded for incomplete hearing data. Ultimately, the study included 2961 subjects for analysis (Figure 1).

Figure 1.

Flowchart of participant inclusion and exclusion for analysis. This flowchart outlines the participant selection process for the study. It shows the initial 31,126 NHANES participants, with exclusions based on age (under 40 or over 69 years old), missing symptomatic dizziness or audiometry data, leaving 2961 participants for final analysis.

2.2. Hearing Measures

Hearing measurements were conducted in a standardized soundproof mobile booth by trained audiologists using American National Standards Institute (ANSI) specifications. Hearing thresholds were measured at frequencies of 500, 1000, 2000, 3000, 4000, 6000, and 8000 Hz using pure tone air conduction audiometry. This included the modified Hughson‐Westlake procedure and the audiometer's automatic testing mode. The speech‐frequency Pure‐Tone Average (PTA) was calculated as the average of hearing thresholds at 500, 1000, 2000, and 4000 Hz. The high‐frequency PTA averaged thresholds at 4000, 6000, and 8000 Hz, and the low‐frequency PTA at 500, 1000, and 2000 Hz. Hearing loss was defined according to the WHO as a speech‐frequency PTA ≥ 20 dB in the better hearing ear [20, 21]. If there are missing values at any of the frequencies (500, 1000, 2000, 4000, 6000, and 8000 Hz) in either ear, the hearing data is considered incomplete and excluded from the analysis.

2.3. Symptomatic Dizziness

Symptomatic dizziness was assessed using NHANES database questions: “During the past 12 months, have you had dizziness, difficulty with balance or difficulty with falling?” A “Yes” response indicated symptomatic dizziness. For evaluating the effectiveness of treatments for symptomatic dizziness, participants who answered “Yes” to “Have you ever been treated by a doctor or other health professional for dizziness, a balance problem, or falling?” were asked: “As a result of this treatment, did your condition get better, get worse, or stay the same?” Outcomes were categorized into “get better,” “get worse,” and “stay the same,” excluding those who refused to answer or stated “don't know.”

2.4. Covariates

The NHANES family interviews gathered comprehensive demographic and health‐related data, including age, gender, race/ethnicity, educational attainment, and family income‐to‐poverty ratio (PIR), among other factors. Body measurements were recorded by trained examiners in the mobile examination center. Tobacco exposure is assessed via serum cotinine levels, alcohol intake is determined by drinking status over the past 12 months, and disease status is based on self‐reporting. In addition, NHANES laboratory data included partial metrics such as complete blood count, biochemical assays, and trace elements.

2.5. Statistical Analysis

The statistical analysis was performed using R (version 4.3.0) and Empower Stats (version 2.0). Sample weights were used to account for the intricate sampling design in accordance with NHANES analysis criteria. The study utilized t‐tests and chi‐square tests to analyze and compare the demographic features of participants who had symptomatic dizziness and those who did not. For the laboratory test results, we applied the Bonferroni correction method to control for multiple comparisons. The original significance level of 0.05 was divided by the total number of tests (55), yielding a corrected significance threshold of 0.00091. This threshold was used for determining statistical significance across all laboratory tests. Logistic regression analysis explored the relationship between average hearing threshold and symptomatic dizziness, calculating odds ratios (ORs) and 95% confidence intervals (CIs). The unadjusted model did not adjust for covariates, while adjusted model 1 accounted for age, gender, and ethnicity, and adjusted model 2 additionally adjusted for educational level, PIR, marital status, tobacco exposure, alcohol intake, BMI, waist circumference, hypertension, diabetes, stroke, headaches/migraines, neck pain, and osteoporosis/brittle bones, anemia, joint pain/aching/stiffness, arthritis, flu/pneumonia/ear infection. Given that elderly individuals frequently encounter various functional impairments associated with hearing loss [22], and that factors such as gender and comorbidities may impact the findings, we performed subgroup and interaction analyses (with respect to comorbidities) to explore the association between hearing loss and dizziness across distinct populations. RCS analysis was used to examine the relationship between hearing levels and symptomatic dizziness. Univariate and multivariate logistic regression models, along with least absolute shrinkage and selection operator (LASSO) regression analysis, identified potential risk factors for hearing loss‐related symptomatic dizziness. Propensity score matching (PSM) was conducted on populations with and without hearing loss, and chi‐square tests assessed the correlation between hearing loss and the effectiveness of symptomatic dizziness treatment. The significance threshold was a two‐tailed p‐value < 0.05.

3. Results

3.1. Baseline Characteristics

This study included 2961 participants, among whom 849 (28.67%) experienced hearing loss, 682 (22.99%) exhibited symptomatic dizziness, and 237 (8.00%) suffered from both conditions. In the non‐hearing loss cohort (speech‐frequency PTA at 10.42 dB), the prevalence of symptomatic dizziness was 21.07%, whereas it escalated to 27.92% in those with hearing loss (speech‐frequency PTA at 29.73 dB) (Table 1, Figure 2A). Statistically significant difference were observed in speech‐frequency PTA among the subgroups: individuals with hearing loss but without symptomatic dizziness (29.27 ± 10.17 dB), those with both conditions (30.96 ± 12.09 dB), participants without either condition (10.33 ± 4.93 dB), and those without hearing loss but with symptomatic dizziness (10.75 ± 4.62 dB). When comparing the group of people who experienced dizziness with those who did not, it was found that the dizziness group was mostly composed of older individuals, and there was a substantially higher percentage of females (63.41%). The average speech‐frequency PTA level in the dizziness group (16.96 dB) was higher than that in the non‐dizziness group (14.28 dB), with similar trends observed in low‐frequency and high‐frequency PTA (Table 1). Additionally, the prevalence of several ailments such as hypertension, diabetes, and headaches/migraines was significantly greater in the group experiencing dizzy symptoms (Supporting Information: Table 1).

Table 1.

The weighted demographic characteristics of study participants by symptomatic dizziness (NHANES 1999–2004 cycle, n = 2961).

Characteristics	Symptomatic dizziness		p value
	No (n = 2279)	Yes (n = 682)
Age, %			0.001
40–49	41.729	33.339
50–59	34.729	38.715
60–69	23.542	27.946
Gender, %			< 0.001
Female	49.119	63.411
Male	50.881	36.589
Speech‐frequency PTA, mean ± SD, dB	14.279 ± 10.000	16.963 ± 12.587	< 0.001
Low‐frequency PTA, mean ± SD, dB	11.233 ± 8.697	14.235 ± 11.553	< 0.001
High‐frequency PTA, mean ± SD, dB	31.376 ± 21.170	35.376 ± 32.328	< 0.001
Ethnicity, %			0.0101
Mexican American	4.926	4.820
Other Hispanic	4.672	6.070
Non‐Hispanic White	77.148	71.116
Non‐Hispanic Black	9.114	11.528
Other race––including multiracial	4.141	6.466
Educational levels, %			< 0.001
Less than high school	16.184	24.259
High school or equivalent	24.545	25.525
College or above	59.271	50.216
Family income‐to‐poverty ratio (PIR), %			< 0.001
< 1.30	12.985	27.481
1.30–3.49	29.145	33.950
≥ 3.50	57.870	38.569
Marital status, %			0.004
Married	69.175	61.911
Widowed	3.279	4.131
Divorced	11.834	16.43
Separated	2.520	2.660
Never married	5.545	6.410
Living with partner	3.605	5.437
Tobacco exposure, %			0.011
Yes	27.851	33.225
No	72.149	66.775
Alcohol intake, %			0.003
Yes	89.977	85.256
No	9.972	14.744
BMI, mean ± SD, kg/m2	28.683 ± 6.078	29.294 ± 7.221	0.034
Waist circumference, mean ± SD, cm	98.831 ± 15.119	99.375 ± 16.250	0.437

Abbreviations: BMI, body mass index; NHANES, National health and nutrition examination survey; PTA, pure‐tone average.

Figure 2.

The hearing level of different populations and the lasso regression results of hl‐sd risk factors. (A) This panel compares the speech‐frequency pure‐tone average (PTA) among four groups: hearing loss with non‐symptomatic dizziness (HL+non‐SD), hearing loss with symptomatic dizziness (HL + SD), nonhearing loss with non‐symptomatic dizziness (non‐HL+non‐SD), and non‐hearing loss with symptomatic dizziness (non‐HL + SD). Statistically significant differences in PTA levels across these groups are indicated. (B) Coefficient profiles from the least absolute shrinkage and selection operator (LASSO) regression, highlighting the factors contributing to HL‐SD risk. (C) The receiver operating characteristic (ROC) curve for the LASSO regression model assessing the predictive power of HL‐SD risk factors, with an AUC of 0.764. (D) Binomial deviance plot of the LASSO model showing the regularization parameter (λ) and the optimal model fit. This analysis identifies key variables contributing to HL‐SD prediction.

Based on the laboratory analysis results, significant differences were observed between the symptomatic dizziness group and the non‐symptomatic dizziness group across several laboratory markers (Supporting Information: Table 2). Specifically, the symptomatic dizziness group showed significantly lower levels of hemoglobin (p = 0.00003) compared to the non‐symptomatic dizziness group, while C‐reactive protein levels were significantly higher (p = 0.00003). Additionally, the symptomatic dizziness group exhibited significantly lower hematocrit levels (p = 0.00006). The concentration of methylmalonic acid was significantly higher in the symptomatic dizziness group (p = 0.00009), as were levels of insulin (p = 0.00013), glucose (p = 0.00017), red cell distribution width (p = 0.00061), and C‐peptide (p = 0.00083). These findings highlight significant differences in several key biochemical markers between the symptomatic and non‐symptomatic dizziness groups, with all differences showing statistical significance after Bonferroni correction.

3.2. Association Between Hearing and Symptomatic Dizziness

In the analyses of unadjusted models, adjust model 1, and adjust model 2, a positive correlation was observed between the prevalence of symptomatic dizziness and PTA at speech, low, and high frequencies. In adjusted model 2, for every 1 dB increase in speech‐frequency and low‐frequency PTA, the risk of symptomatic dizziness increases by 1.6% and 1.7%. For high‐frequency PTA, each 1 dB increment is associated with a 0.8% increase in risk (Table 2).

Table 2.

The association of threshold of hearing with the prevalence of symptomatic dizziness from NHANES 1999 to 2004.

Characteristics	Unadjusted model	Adjusted Model 1a	Adjusted Model 2b
	OR (95% CI), p‐value	OR (95% CI), p‐value	OR (95% CI), p‐value
Speech‐frequency PTA (dB)	1.017 (1.010, 1.025), < 0.001	1.024 (1.015, 1.032), < 0.001	1.016 (1.005, 1.027), 0.003
Low‐frequency PTA (dB)	1.024 (1.016, 1.032), < 0.001	1.025 (1.017, 1.034), < 0.001	1.017 (1.006, 1.029), 0.003
High‐frequency PTA (dB)	1.005 (1.002, 1.008), 0.003	1.010 (1.005, 1.015), < 0.001	1.008 (1.001, 1.014), 0.023
Grouping of speech‐frequency PTA (dB)
PTA < 20 dB	Reference	Reference	Reference
20 dB ≤ PTA < 35 dB	1.366 (1.118, 1.671), 0.002	1.492 (1.197, 1.859), < 0.001	1.312 (0.996, 1.729), 0.054
35 dB ≤ PTA < 50 dB	1.629 (1.108, 2.395), 0.013	1.933 (1.286, 2.905), 0.002	1.822 (1.096, 3.028), 0.021
50 dB ≤ PTA < 65 dB	2.029 (1.025, 4.017), 0.042	2.535 (1.265, 5.081), 0.009	2.079 (0.800, 5.400), 0.133
65 dB ≤ PTA < 80 dB	1.873 (0.467, 7.519), 0.376	2.334 (0.548, 9.946), 0.252	1.976 (0.371, 10.509), 0.424
PTA ≥ 80 dB	5.619 (0.936, 33.732), 0.059	5.604 (0.864, 36.370), 0.071	1.777 (0.200, 15.750), 0.021
p for trend	< 0.001	< 0.001	0.004

Abbreviations: CI, confidence interval; dB, decibel; NHANES, National Health and Nutrition Examination Survey; OR, odd ratio; PTA, pure‐tone average.

^a Adjusted model 1: adjust: gender, age, and ethnicity.

^b Adjusted model 2: adjust: gender, age, ethnicity, educational level, family income‐to‐poverty ratio (PIR), marital status, tobacco exposure, alcohol intake, noise exposure, BMI, waist circumference, hypertension, diabetes, stroke, headaches/migraines, neck pain, osteoporosis/brittle bones, anemia, joint pain/aching/stiffness, arthritis, flu/pneumonia/ear infection.

Additional sensitivity analysis was performed to strengthen the reliability of the findings. The study classified speech‐frequency hearing loss into six categories according to the severity levels of hearing impairment as defined by the WHO. These categories range from normal hearing to mild, moderate, moderately severe, severe, and profound or higher. Individuals with varied degrees of hearing loss exhibited a greater propensity for symptomatic dizziness compared to those with normal hearing. As the severity of hearing loss increases, there is a statistically significant trend of acquiring symptomatic dizziness (p for trend < 0.01) (Table 2).

Subgroup analysis confirmed the strong connection between hearing threshold levels and the frequency of clinical dizziness (Figure 3). Both males and females showed a positive link between speech, low, and high‐frequency PTA, and the rate of symptomatic dizziness, with a somewhat stronger correlation in females. The 50–59 age group had a more pronounced positive connection between speech‐frequency and low‐frequency PTA with the occurrence rate of symptomatic dizziness, which appeared to decline in the 60–69 age group. Non‐Hispanic whites and other Hispanics demonstrated notable positive associations between the different frequencies of PTA and the prevalence of symptomatic dizziness. This link was particularly pronounced in patients with an education level beyond high school. The lower‐income groups showed a clear positive relationship between low‐frequency PTA and symptomatic dizziness. However, this relationship was not statistically significant in higher‐income groups. BMI ≥ 30 kg/m² individuals showed a notable positive correlation across speech, low, and high‐frequency PTA with the prevalence of symptomatic dizziness. Both alcohol intake and tobacco exposure affected the connection between different frequencies of PTAs and the rate of symptomatic dizziness.

Figure 3.

The association between hearing threshold and symptomatic dizziness by selected subgroups. This figure visualizes the correlation between hearing thresholds (speech‐frequency, low‐frequency, and high‐frequency PTA) and symptomatic dizziness across various subgroups. Subgroups include gender, age, ethnicity, educational level, body mass index (BMI), alcohol intake, tobacco exposure, and family income‐to‐poverty ratio (PIR). The plot uses forest plots to represent the odds ratios (OR) of symptomatic dizziness with increasing hearing thresholds for each subgroup.

We assessed potential effect modification by testing multiplicative interaction terms between speech‐frequency PTA and nine major comorbidities: hypertension (p = 0.44), diabetes (p = 0.07), joint pain (p = 0.39), arthritis (p = 0.47), osteoporosis (p = 0.48), headaches/migraines (p = 0.70), neck pain (p = 0.39), flu/pneumonia/ear infection (p = 0.27), anemia (p = 0.47) and stroke (p = 0.79) (Figure 4, Supporting Information: Table 3). The nonsignificant interaction terms indicate that the association between hearing loss and symptomatic dizziness persists across these comorbid conditions without statistically detectable modification effects. This stability of the primary association in stratified analyses suggests that hearing loss may exert risk effects through pathways distinct from these comorbidities, though unmeasured biological interactions cannot be excluded.

Figure 4.

Interaction analyses between speech‐frequency pure‐tone average (PTA) and comorbidities on symptomatic dizziness. This figure shows interaction analyses between speech‐frequency PTA and comorbidities (e.g., hypertension, joint pain, arthritis, headaches) and their effect on the prevalence of symptomatic dizziness. Each plot presents the relationship between PTA and dizziness prevalence stratified by the presence of these comorbidities. The shading represents confidence intervals, while the color lines distinguish between conditions, showing how comorbidities modulate the impact of hearing loss on dizziness.

The results indicated that using RCS, a non‐linear fitting model, a significant association was observed between low‐frequency PTA and symptomatic dizziness in both unadjusted and fully adjusted models, with no significant nonlinear trend. For speech‐frequency PTA, a significant association with symptomatic dizziness was found in both models, with a non‐linear relationship present in the unadjusted model. However, this non‐linear trend disappeared after full adjustment for covariates. No significant relationship was observed between high‐frequency PTA and symptomatic dizziness (Figure 5).

Figure 5.

The restricted cubic spline (RCS) of association between low frequency PTA, speech frequency PTA, high frequency PTA, and symptomatic dizziness. This figure illustrates the non‐linear relationships between different frequencies of PTA (low‐frequency, speech‐frequency, and high‐frequency) and symptomatic dizziness using restricted cubic splines (RCS). The plots show that as hearing thresholds increase, the risk of dizziness also increases, with speech‐frequency PTA exhibiting the strongest non‐linear association. After adjusting for covariates, the non‐linear trend disappears. The shaded areas in the figure represent the 95% confidence intervals.

3.3. Risk Factors for Hl‐SD

Univariate and multivariate logistic regression analyses were conducted on individuals with hearing loss (speech‐frequency PTA ≥ 20 dB) (Supporting Information: Table 4). Univariate analysis indicated that several demographic factors and health conditions were significantly linked with HL‐SD. Multivariate analysis identified females (OR = 2.34, p < 0.01), living with a partner (OR = 2.76, p = 0.03), BMI (OR = 0.92, p = 0.04), stoke (OR = 1.75, p = 0.01), joint pain/aching/stiffness (OR = 2.38, p < 0.01), flu/pneumonia/ear infection diabetes (OR = 1.73, p < 0.01), diabetes (OR = 2.32, p < 0.01), and tinnitus (OR = 2.32, p < 0.01) as significant predictors of HL‐SD. LASSO regression identified 11 key predictors of HL‐SD, including gender, PIR, arthritis, stroke, neck pain, joint pain/aching/stiffness, hypertension, diabetes, headaches/migraines, tinnitus, and insulin levels (Figure 2B–D, Supporting Information: Table 5). The regularization parameter λ was 0.04, and the area under the curve (AUC) of the Receiver Operating Characteristic (ROC) curve was 0.76 (Figure 2C).

3.4. Impact of Hearing Loss on the Therapeutic Effectiveness of Symptomatic Dizziness

We analyzed the impact of hearing levels on the prognosis of symptomatic dizziness in a sample of 213 adults selected by questionnaires. Employing a 1:1 PSM technique that takes into account gender, age, and ethnicity, the results indicated that there were no statistically significant disparities (p = 0.38) in the prediction of symptomatic dizziness between those with hearing loss (n = 63) and those with normal hearing (n = 63) (Supporting Information: Table 6).

4. Discussion

This cross‐sectional study involved 2,961 participants explored the comorbidity patterns between HL‐SD. Our findings show a significant trend: as hearing loss worsens, the incidence of symptomatic dizziness increases. For each 1 dB increase in low‐frequency and mid‐frequency PTA, the risk of symptomatic dizziness escalates by 1.6%. Similarly, a 1 dB increase in high‐frequency PTA results in a 0.7% rise in risk. Further explorations through univariate and multivariate regression analyses identified risk factors for HL‐SD. We also developed a comorbidity risk model using results from multiple laboratory tests combined with previously analyzed demographic and clinical variables. Based on LASSO regression, this model highlights that gender, PIR, arthritis, stroke, neck pain, joint pain/aching/stiffness, hypertension, diabetes, headaches/migraines, tinnitus, and insulin levels are closely linked with HD‐SL. While hearing loss significantly correlates with symptomatic dizziness, it does not influence the treatment outcomes for this condition.

The co‐morbidity of hearing loss and dizziness is highlighted in several disease guidelines, which discuss the characteristics and epidemiology [14, 15, 23, 24]. The 2019 updated American Clinical Practice Guideline: Sudden Hearing Loss indicates that 30%–60% of Sudden Sensorineural Hearing Loss (SSNHL) cases are accompanied by dizziness, adding significant disease burden for patients with SSNHL. Additionally, the guideline on Clinical practice: Vestibular neuritis notes that acute peripheral vestibular dysfunction syndromes, such as vestibular neuritis or labyrinthitis, often present with sudden, severe vertigo with a sensation of rotational motion, and may also include hearing loss and tinnitus if the auditory part of the inner ear is involved [24]. The Clinical Practice Guideline: Ménière's Disease reports that most patients with MD experience some degree of hearing loss, tinnitus, aural fullness, or balance disorders, with nearly one‐third suffering from severe symptoms in at least one of these areas, and 82% experiencing moderate to severe hearing loss (speech‐frequency PTA > 50 dB) [14]. Earlier studies have explored the incidence of HL‐SD in specific diseases, while our study quantifies the association between the two and identifies comorbid risk factors, thereby enriching the understanding of their comorbidity patterns and offering significant research value.

Dizziness with symptoms may be caused by several conditions affecting multiple organs and systems, such as benign paroxysmal positional vertigo, MD, migraine, acute peripheral vestibulopathy, cerebral ischemia, and anxiety disorders. Of all these diseases, those that impact vestibular function are especially closely linked to hearing loss [25, 26, 27, 28]. The susceptibility to vestibular dysfunction and hearing loss may be due to the close anatomical proximity of the vestibular apparatus and cochlea [29], as well as their shared blood supply [30]. This makes both structures vulnerable to degenerative, ischemic, traumatic, or toxic damage [31, 32], as well as various factors such as inflammation, metabolism, immunity, and infection [33, 34, 35, 36]. There are other factors that can elevate the risk of HL‐SD. Microcirculatory abnormalities and inadequate blood supply to the ear are major contributing factors. The inner ear necessitates a consistent blood flow to preserve regular auditory function and equilibrium. Diabetes has the potential to impair the construction and function of micro vessels, cause damage to the spiral ganglion neurons, afferent nerve fibers, the organ of Corti, and the stria vascularis of the inner ear, resulting in hearing loss [37, 38]. Elevated blood glucose levels can result in ischemic alterations in the vestibular organs and peripheral neuropathy via microangiopathy, resulting in symptoms of dizziness and a sensation of imbalance [39, 40, 41, 42]. Hypertension speeds up the progression of vascular sclerosis, which impacts the flow of blood and can result in reduced blood supply to the vestibular labyrinth, causing symptoms related to balance and spatial orientation [30, 43, 44]. Additionally, it can disrupt the potassium ion cycle in the inner ear, decrease the oxygen partial pressure in the cochlea, leading to impairment of the vascular stria and impacting hearing ability [45, 46]. The neck serves as a vital arterial conduit, via which the carotid and vertebral arteries supply blood to the brain and other organs in the head region. Neck and joint discomfort or stiffness can occur due to muscle stress, injury, or degenerative conditions that cause nerve compression and insufficient blood flow, which may eventually develop to HL‐SD [47, 48, 49].

Other biological mechanisms, such as abnormalities in the central nervous system, hormone levels, and psychosocial factors, may also contribute to HL‐SD. Studies suggest that migraine patients often experience symptoms related to audio vestibular dysfunction, such as vertigo, tinnitus, and hearing loss [50, 51, 52]. These symptoms are probably linked to neuroinflammation in the trigeminal vascular system. The trigeminal nerve facilitates pain signal transmission through the release of neuropeptides, including calcitonin gene‐related peptide (CGRP), substance P, neurokinin A, and nitric oxide. These substances result in vasodilation, mast cell degranulation, and neurogenic inflammation. CGRP, a neuropeptide, plays a role in the efferent synapses of hair cell organs, such as the cochlea, semicircular canals, and lateral line system. Animal studies have shown that a CGRP deficiency correlates with reduced suprathreshold cochlear nerve activity and a lower vestibulo‐ocular reflex (VOR) gain. Additionally, mice lacking CGRP also show impaired otoconial activity and diminished balance capabilities. Research indicates that women may be more vulnerable to hormonal fluctuations; for instance, changes in progesterone and estrogen levels can influence the balance of inner ear fluids and auditory function [53, 54, 55]. Furthermore, women and those living with partners are more likely to report these symptoms, possibly due to psychological factors. Patients with tinnitus face a higher risk of HL‐SD, likely due to a combination of the aforementioned factors. Tinnitus and hearing loss are generally associated with damage to inner ear hair cells, caused by noise exposure, aging, drug toxicity, infections, inadequate blood supply, or other causes [56, 57, 58]. Damage to the inner ear can also impair the balance system, leading to dizziness. Moreover, hearing loss and tinnitus may increase anxiety, depression, and psychological stress in patients [59, 60, 61], which could indirectly worsen the severity of dizziness [62, 63, 64].

The effectiveness of dizziness treatment can depend on several factors, such as the vestibular system's plasticity, individual adaptability, and overall health status [28, 65, 66, 67]. Therefore, even with severe hearing loss, other prognostic factors may have a greater impact on both the immediate and long‐term outcomes of dizziness treatment, making hearing loss not a critical determinant.

4.1. Limitations

Firstly, the cross‐sectional design of the NHANES dataset precludes determining the onset and progression of hearing loss and dizziness, potentially introducing bias and complicating causality analysis. Future studies should explore their causal relationship. Second, despite adjusting for many confounding factors including demographics and medical comorbidities, residual confounding from unmeasured variables such as hereditary hearing loss cannot be excluded. Notably, genetic variations including point mutations and copy number variations may differentially influence inner ear pathophysiology [68, 69, 70, 71]. The absence of genetic profiling may affect the generalizability of observed comorbidity patterns given these innate phenotypic variations. Another limitation of this study is the lack of mental health data, such as depression and anxiety, which prevented further adjustment for these potential confounders. Additionally, using self‐reported dizziness symptoms offers valuable insights but is subject to individual biases in interpretation and recall.

5. Conclusions

Our study confirms a significant correlation between hearing loss and increased symptomatic dizziness, with risk rising by 1.6% for every 1 dB increase speech‐frequency PTA. We identified key risk factors such as headache, diabetes, and tinnitus for this comorbidity. Despite the strong association, hearing loss did not affect dizziness treatment outcomes. This highlights the need for targeted interventions and further research to manage and understand these complex interactions effectively.

Author Contributions

Chunyan Liu was responsible for the conceptualization of the study and drafting the initial manuscript. Xiaonan Wu conducted formal analysis and created visualizations. Jin Li contributed to methodology design and validation processes. Jiao Zhang managed the project and provided supervision. Yun Gao also supervised the work and revised the manuscript during the review and editing stages. Da‐Yong Wang was involved in reviewing and refining the manuscript, while Qiuju Wang secured funding for the project and contributed to the final review and editing.

Ethics Statement

This study was approved by NCHS Ethics Review Board (Protocol #98‐12).

Conflicts of Interest

Professor Qiu‐Ju Wang is a member of World Journal of Otorhinolaryngology––Head and Neck Surgery (WJOHNS) editorial board and is not involved in the peer review process of this article.

Supporting information

Supplementary Table 1: The weighted characteristics of other medical conditions in study participants with symptomatic dizziness. Supplementary Table 2: Comparison of mean laboratory values between symptomatic dizziness and non‐symptomatic dizziness groups. Supplementary Table 3: Association of hearing loss and comorbidities with dizziness. Supplementary Table 4: Analysis of risk factors for symptomatic dizziness and hearing loss using univariate and multivariate logistic regression models. Supplementary Table 5: Coefficients of Risk Factors for HL‐SD Derived from LASSO Regression. Supplementary Table 6: Comparison of basic characteristics and treatment effect on symptomatic dizziness between matched groups.

Liu C.‐Y., Wu X.‐N., Li J., et al., “Comorbidity Patterns Between Hearing Loss and Symptomatic Dizziness in Middle‐Aged and Older Adults,” World Journal of Otorhinolaryngology ‐ Head and Neck Surgery 0 (2025): 1‐12, 10.1002/wjo2.70078.

Data Availability Statement

All data generated or analyzed during this study are included in the additional files for review. Detailed raw data files are available from the corresponding author on reasonable request.