Development and Application of a Reference-Interval Approach to Tympanometric Norms Using U.S. Population Data for...

United States national data for tympanometric measures from 16,614 ears included in the National Health and Nutrition Examination Survey for 2015-2016 and 2017-2020 were analyzed and described.

Analyses were primarily descriptive and relied on distribution-free medians and cumulative distribution functions (CDFs). Differences between CDFs were examined using Cohen's h effect size. Results are presented for the age range 6-80+ years. A reference interval (RI) approach was pursued in which the interval from the 2.5th to the 95th percentile for healthy ears was specified for peak-compensated static admittance (Y_tm+), tympanometric peak pressure, tympanometric width, and equivalent acoustic ear-canal volume (V_ea+).

Separate RIs were required for 6- to 13-year-olds compared to those 14-80+ years of age for all four tympanometric measures. For both age groups, a separate RI was warranted for each sex for Y_tm+ and V_ea+. For those 14-80+ years of age, a separate RI was also needed for non-Hispanic Asians for Y_tm+. When these RIs were applied to the full data set, about 5%-10% of ears were found to be either below the lower limit or above the upper limit of the RI. Those ears outside the RI limits had about 5 dB higher low-frequency hearing thresholds than those within RI limits. Both Y_tm+ and V_ea+ were found to depend on body size.

RIs for tympanometric measures from individuals with healthy ears were established, and when applied to the general U.S. population 6-80+ years of age, 5%-10% of ears fell outside the RIs depending on the measure.