The Cochlear Center has led the integration of hearing measures into numerous large epidemiological studies including the Atherosclerosis Risk in Communities (ARIC) Study and the NIA’s Baltimore Longitudinal Study of Aging. Developing a strong protocol requires ample prior planning and time investment for training. There is a wide range of possible hearing measures from using simple self-report measures to collecting detailed objective measures with pure-tone audiometry and speech tests. Myriad factors need to be considered when developing an appropriate protocol including: what the hearing measurement is truly assessing with respect to hearing and communicative function, the sensitivity and accuracy of the measurement, time requirements for testing, equipment costs, space/environment factors, calibration schedules, staff training time, and the exposure/outcome of interest.
See protocols samples that we have used for various studies (BHMH in Yellow and ARIC in Purple).
Self-report binary (yes or no) measures have been found to be inaccurate or insensitive to detecting hearing loss. However, using subjective questions with ordinal scales (i.e., including different degrees of hearing loss) and taking into account demographic information such as age, race and sex can greatly improve the sensitivity of such simple measures when compared to gold-standard audiometry. Center Core Faculty Joshua Betz is currently developing measurement error models incorporating basic demographic information to allow subjective measures to have greater predictive accuracy when compared to audiometry. Such approaches are essential when time/resources/priorities only allow a few subjective questions on hearing to be included in large studies.
Beyond the equipment and protocol selection, quality is assured via training, continuous data monitoring and equipment upkeep. Comprehensive staff training with skills assessment over multiple time points for certification during the study is key to setting up a study for success. Further, regular data monitoring should be completed to look for abnormalities in the audiometric data based on population and biologic norms. Detected abnormalities may reveal issues in protocol adherence and/or equipment malfunction.
For questions or help regarding protocol development and including assessments of hearing into clinical or epidemiological research studies, please contact Nicholas Reed at firstname.lastname@example.org. Funding support is also available to researchers interested in incorporating measures of hearing into their studies.