Journal Logo


NHANES: No significant link between cardiorespiratory fitness and tinnitus

Loprinzi, Paul D. PhD; Cardinal, Bradley J. PhD; Gilham, Ben

Author Information
doi: 10.1097/
  • Free

Other studies have pointed to a link between cardiovascular fitness and hearing sensitivity, but our review of the National Health and Nutrition Examination Survey (NHANES) found no significant correlation between a person's physical condition and tinnitus.

Although 51% of men without tinnitus were highly fit compared with 44% of highly fit male participants with tinnitus, the overall findings do not provide convincing evidence that cardiovascular fitness is associated with tinnitus. It is difficult, however, to compare our findings with previous studies because no other research has examined this association. Our findings, although not a direct comparison, are similar to the Pratt et al1 study that showed cardiovascular health was not associated with hearing levels in 548 participants between 72 and 96 years old.

As hearing healthcare professionals know, hearing loss is the third most common chronic disease occurring among adults.2 The prevalence of hearing impairment among adults 65 to 69 is most pronounced at 40% to 45%.3 Researchers have hypothesized that cardiovascular disease, in addition to genetics, age, exposure to loud noise, and ototoxic medications, may also play a role in hearing function.4 Previous research has revealed that cardiovascular fitness has a protective effect on hearing function,5–8 with hypothesized mechanisms including increased blood flow, reduced blood pressure, and oxygen delivery to the cochlea.6,8 Hutchinson et al showed last year that older adults with low cardiovascular fitness had significantly worse pure-tone hearing at 2000 and 4000 Hz, compared with those at medium and high fitness levels.6

The possibility that cardiovascular fitness is associated with hearing function, specifically hearing sensitivity, makes it vital for public health to determine if cardiovascular fitness is associated with other hearing-related conditions, such as tinnitus, especially because recent estimates indicate that 26% to 30% of individuals experience this condition.9 This study examined the association between cardiorespiratory fitness (i.e., maximum oxygen consumption [VO2max]) and tinnitus among a nationally representative sample of adults to address this gap in the research literature.


Design and participants

Data from the 1999-2000, 2001-2002, and 2003-2004 NHANES cycles were combined to increase the sample size and to produce estimates with greater statistical reliability. Participants were interviewed in their homes and later examined in mobile examination centers across 15 US geographic locations during each of the three cycles. The study was approved by the National Center for Health Statistics ethics review board, with informed consent obtained from all participants prior to data collection. The final sample of our study included 2,303 NHANES participants. Participants were excluded if their VO2max data were missing, if they self-reported tinnitus, or if they reported being pregnant, being exposed to firearm noise exposure, or had been exposed to loud noise outside of work.

Measurement of VO2max

The cardiorespiratory fitness of eligible participants aged 12 to 49 was assessed on a treadmill at the mobile examination centers. Participants were excluded by certain medical conditions (e.g., previously diagnosed with coronary heart disease or self-reported heart problems), medications (e.g., lidocaine), physical limitations (e.g., difficulty walking up 10 steps without resting), limits on resting heart rate (i.e., more than 100 beats per minute) and blood pressure (systolic blood pressure higher than 180 mm Hg; diastolic blood pressure higher than 100 beats per minute), irregular resting heart rates (more than three beats per minute), and other reasons specified by the participant, physician, or staff (e.g., hospitalized in the previous three months).

The submaximal treadmill test was performed by trained health technicians, and participants were assigned to one of eight treadmill test protocols based on gender, age, body mass index, and self-reported levels of physical activity. Each protocol sought to elicit a heart rate approximately 75 percent of the participant's age-predicted maximum heart rate (220-age) by the conclusion of the test.

Each treadmill protocol included a two-minute warmup, two three-minute exercise stages, and a two-minute cool-down. The heart rate was measured continuously using an automated monitor with four electrodes connected to the thorax and abdomen of the participant. Recordings were taken at the end of the warmup, each exercise stage, and each minute of recovery. VO2max was estimated by measuring the heart rate response to known levels of submaximal work because the relationship between heart rate and oxygen consumption is assumed to be linear during exercise.10

The estimated VO2max was categorized as low, moderate, or high cardiovascular fitness based on gender and age-specific criteria. Using definitions from the Aerobics Center Longitudinal Study,10,11 low-level cardiovascular fitness was established as an estimated VO2max below the 20th percentile of the same gender and age group. Moderate fitness was defined as a value between the 20th and 59th percentile, and high fitness was classified as at or above the 60th percentile.

Measurement of tinnitus

In the participants' home, those 20 years and older were asked about their tinnitus status. In the NHANES 1999-2004 cycles, participants were asked if they ever had ringing, roaring, or buzzing in their ears over the past 12 months. Possible responses were yes, no, don't know, and refused to answer. Of the 296 participants who answered affirmatively, 293 in the analyzed sample reported the frequency or severity of tinnitus. These participants were asked how often ringing, roaring, or buzzing occurred, with possible responses including almost always, at least once a day, at least once a week, at least once a month, or less than once a month.

Other measurements

Information about age, gender, race-ethnicity, marital status, education, and exposure to noise outside of work were obtained from a questionnaire administered by trained household interviewers. Data were recorded using a Blaise format computer-assisted personal interview system. Body mass index was also calculated from weight and height measured during the examination at the mobile examination center.

Data Analysis

All statistical analyses were performed using procedures from sample survey data using STATA (version 10.0, College Station, TX) to account for the complex survey design used in NHANES. All analyses included the use of appropriate survey sample weights to account for oversampling and non-responses and to provide nationally representative estimates. When an analysis resulted in a stratum with a single cluster, the variance contribution was centered at the overall cluster mean.

Means and standard errors were calculated for continuous variables, and proportions were calculated for categorical variables. Statistical differences between continuous variables and categorical variables were tested using an adjusted Wald test, a survey-data analog to the parametric t-test (e.g., VO2max and tinnitus status). Statistical differences between categorical variables were tested with design-based likelihood ratio tests (e.g., between tinnitus status and fitness classification).

A logistic regression was used to examine the association between VO2max and tinnitus. The dependent variable (i.e., tinnitus status) was coded as 0 for those having tinnitus and 1 for those not having it. Those having tinnitus served as the reference group. Individuals with low cardiovascular fitness were coded as 1, medium fitness was classified as 2, and high fitness was coded as 3. Logistic regression models were computed for the entire sample and separately for male and female participants. All logistic regression models were adjusted for potential confounding variables, including education, age, race-ethnicity, gender (only in the model for the entire sample), marital status, and BMI.


Table 1 displays the demographic characteristics of the analyzed sample stratified by tinnitus status. No significant differences existed between those with and without tinnitus for demographic variables. Overall, 12.8% of the nationally representative sample reported having tinnitus (296/2303).

Table 1
Table 1:
Weighted means and percentages (standard error) for selected characteristics of the NHANES 1999-2004 samples.

The severity of tinnitus among those who reported having the condition is displayed in Table 2. The majority of participants with tinnitus (53.5%) reported having ringing, roaring, or buzzing in their ears less than once a month.

Table 2
Table 2:
Severity of tinnitus among those who self-reported the condition, NHANES 1999-2004 (n=293).

As expected, significant differences in VO2max occurred across the fitness levels. The mean VO2max for the low, moderate, and high fitness groups was 28.6, 34.8, and 45.3 mL/kg/min, respectively, for male and female participants combined. For male participants, mean VO2max was 32.0, 38.7, and 49.7 mL/kg/min, respectively, for the low, moderate, and high fitness groups. The average VO2max for female participants was 26.1, 31.7, and 41.8 mL/kg/min, respectively, for the low, moderate, and high fitness groups.

Table 3 displays the mean VO2max and percentage in each fitness classification across tinnitus status and gender. No differences in VO2max or fitness level across tinnitus status were observed. Table 4 shows the VO2max for each level of tinnitus frequency among participants who self-reported tinnitus, and results are shown for male and female participants together and separately. When analyzing gender together and separately, there were no differences in VO2max across the different levels of tinnitus frequency. Table 5 shows the proportion in each fitness classification for each level of tinnitus frequency. No differences in the proportion of each fitness classification were seen across the different levels of tinnitus frequency.

Table 3
Table 3:
Weighted mean (standard error) maximum oxygen consumption (VO2max) and fitness classification across tinnitus status and by gender, NHANES 1999-2004.
Table 4
Table 4:
Weighted mean (standard error) maximum oxygen consumption (VO2max) for each level of tinnitus frequency among those participants who self-reported having tinnitus, NHANES 1999-2004.
Table 5
Table 5:
Weighted proportion in each fitness level for each level of tinnitus frequency among those participants who self-reported having tinnitus, NHANES 1999-2004.

The adjusted logistic regression examining the odds of not having tinnitus for every 1-unit increase in VO2max and across fitness classification is displayed in Table 6. After adjusting for age, body mass index, gender, marital status, race-ethnicity, and education, no statistically significant associations emerged for the models computed for the entire sample or separately for male and female participants. In addition to the adjusted logistic regression models, simplified models were computed through a backwards elimination procedure to reduce the model to include only statistically significant factors. The results of these reduced models were not substantially different from the adjusted models, and these data are not shown.

Table 6
Table 6:
Adjusted logistic regression examining the odds of not having tinnitus across fitness and for every unit increase in maximum oxygen consumption (VO2max), NHANES 1999-2004.


The present findings do not provide convincing evidence that cardiovascular fitness is associated with tinnitus. In the adjusted logistic regression for men, the odds ratio for the fitness classification approached statistical significance (odds ratio=1.30), suggesting that men who are highly fit (VO2max ≥ 44.23 mL/kg/min for 20- to 29-year-olds; ≥ 42.42 mL/kg/min for 30- to 39-year-olds; and ≥ 39.89 mL/kg/min for 40- to 49-year-olds) may be less likely to develop tinnitus.

Our null findings may be partly attributed to age restrictions: only participants up to age 49 were included in the cardiorespiratory fitness component of the mobile examination. It is possible that older adults, particularly older men, who more commonly report tinnitus, might exhibit a greater association between the condition and cardiorespiratory fitness. Some 15,279 adults aged 20 to 85 in the 1999-2004 NHANES cycles answered the tinnitus question. Twenty-two percent of adults aged 20 to 49 self-reported tinnitus symptoms while 30.2% of adults 50 to 85 years self-reported the condition.

Future studies should investigate the association between cardiorespiratory fitness and tinnitus in older adults and assess its severity based on the duration of the symptoms during a single period of time.


1. Pratt SR, Kuller L, et al: Prevalence of hearing loss in black and white elders: results of the Cardiovascular Health Study. J Speech Lang Hear Res 2009;52(4):973–989.
2. Lethbridge-Cejku M, Schiller JS, et al: Summary health statistics for U.S. adults: National Health Interview Survey, 2002. Vital Health Stat 10 2004(222):1–151.
3. Cruickshanks KJ, Wiley TL, et al: Prevalence of hearing loss in older adults in Beaver Dam, Wisconsin. The Epidemiology of Hearing Loss Study. Am J Epidemiol 1998;148(9): 879–886.
4. Johnsson LG, Hawkins JE, Jr: Sensory and neural degeneration with aging, as seen in microdissections of the human inner ear. Ann Otol Rhinol Laryngol 1972;81(2):179–193.
5. Alessio HM, Hutchinson KM, et al: Study finds higher cardiovascular fitness associated with greater hearing acuity. Hear J 2002;55(8):32–40.
6. Hutchinson KM, Alessio HM, et al: Association between cardiovascular health and hearing function: Pure-tone and distortion product otoacoustic emission measures. Am J Audiol 2010;19:26–35.
7. Ismail AH, Corrigan DL, et al: Biophysiological and audiological variables in adults. Arch Otolaryngol 1973;97(6):447–451.
8. Manson J, Alessio HM, et al: Does cardiovascular health mediate hearing ability. Med Sci Sports Exerc 1994;26(7):866–871.
9. Hasson D, Theorell T, et al: Prevalence and characteristics of hearing problems in a working and non-working Swedish population. J Epidemiol Community Health. 2010;64(5):453–460.
10. American College of Sports Medicine. ACSM's Guidelines for Exercise Testing and Prescription, 6th edition. Philadelphia, PA: Lippincott Williams and Wilkins, 1995.
11. Blair SN, Kohl HW 3rd, et al: Physical fitness and all-cause mortality. A prospective study of healthy men and women. JAMA 1989;262(17):2395–2401.
© 2011 Lippincott Williams & Wilkins, Inc.