Patterns of quality in audiometric data for U.S. workers.
Spectrum 2013 Apr; 30(1):18-19
In the last couple of years, the National Institute for Occupational Safety and Health (NIOSH) Occupational Hearing Loss (OHL) Surveillance Project team has analyzed audiometric data from large numbers of U.S. workers. For those unfamiliar with the Project, NIOSH partnered with audiometric service providers who shared millions of de-identified audiograms and demographic data with NIOSH for OHL surveillance and research. Some interesting patterns related to quality have emerged. The quality of the audiograms conducted, and the completeness of the demographic information collected at the time of testing, is critical. It is essential for successful hearing loss prevention efforts for the individual worker, and for valid and reliable research results to be obtained. In this article, we would like to explain the methodology used to identify audiograms with quality issues and briefly provide the percentages of excluded audiograms from our latest analysis. Our current quality assurance methods were developed over time, beginning in the 1990s, by senior audiologists within the NIOSH Hearing Loss Prevention Team as they analyzed various audiometric datasets collected for regulatory rather than research purposes. They designed these methods to identify audiograms with quality deficiencies that could indicate inaccurate thresholds, or audiometric patterns consistent with non-occupational exposures or pathology. We applied these methods when analyzing our Project's large audiometric database and published the methods in 20121. This article focuses only on quality deficiencies, including missing information and suspect thresholds.
Hearing-loss; Hearing-tests; Audiometers; Workers; Analytical-processes; Demographic-characteristics; Hearing-protection; Exposure-levels; Pathology; Surveillance-programs
Newsletter; Lay Publication
Spectrum: the National Hearing Conservation Association newsletter