Tinnitus Screener: Short-Term Test-Retest Reliability.

Abstract

The Tinnitus Screener was introduced in 2015 as a four-item algorithmic instrument to assess the temporal characteristics of a person's reported tinnitus. The Tinnitus Screener was then revised as a six-item version to include a new temporal category and to capture tinnitus duration (acute < 6 months vs. chronic ≥ 6 months). When contrasted with audiologist assessment, the four-item Tinnitus Screener was determined to be highly valid, but the short-term reliability of either version remained unknown. The present analysis focused on determining the test-retest reliability of the six-item Tinnitus Screener. Additionally, we sought to determine whether reliability differed by respondent age, sex, military status, and hearing loss. The Tinnitus Screener was administered to 190 military Service members and 250 military Veterans at two time points separated by 7-31 days. Our analysis focused on test-retest reliability of responses as measured by the kappa coefficient, overall and within subsamples. Percent agreement of tinnitus categorization (temporal categories) and classification (positive/negative) between the two time points was also evaluated. Constant or intermittent tinnitus was found in 31% of Service members and 53% of Veterans. Overall, kappa reliability coefficients were high, near .80, indicating substantial reliability. The majority (96%) of reliability coefficients for the Tinnitus Screener within subsamples were similarly high, ranging from .68 to .88. The updated version of the Tinnitus Screener is shown to be a reliable instrument. The Tinnitus Screener is recommended to inform clinical decision making by determining the temporal characteristics of tinnitus.