Test-retest reproducibility of the 1000 Hz tympanometry test in newborn and six-week-old healthy infants.

Abstract

This study aimed to evaluate the test-retest reproducibility of the high frequency tympanometry (HFT) test measured in healthy infants. A total of 273 newborn babies (106 males and 147 females) were assessed twice (Test 1 and Test 2) on the same day, followed by two more assessments (Test 3 and Test 4) for 118 babies (48 males and 70 females) who returned six weeks later. Five HFT measures including the peak compensated static admittance and component compensated static admittance were assessed for test-retest reproducibility. The results showed no significant differences in mean values of the HFT results between the test and retest conditions for newborn (Test 1/ Test 2) and six-week-old babies (Test 3/ Test 4). High reproducibility for all HFT measures was found for both age groups, as judged by the high intra-correlation coefficients of between 0.75 and 0.95. Normal variations of the HFT measures were established using the 90% range of absolute test-retest difference. Changes in test-retest findings exceeding the 95th percentile values may be considered significant, indicating possible functional changes.