Clinical trial of BOBCAT: 1st report on the reliability and validity of computerized pure-tone audiometry

Computerized diagnostic audiometry is quickly emerging as a viable productivity tool in the audiology clinic. To date, there has been little reported on its reliability and validity with the hearing-impaired. The 'Battery of Basic Computerized Audiometric Tests' (BOBCAT) is a computer program which puts a wide variety of clinical hearing tests under computer control. The purpose of the present study was to ascertain the reliability and validity of BOBCAT in the measurement of hearing sensitivity. A field study was conducted in a group of 92 workers exposed to noise to measure air- and bone-conduction thresholds. Coefficients of reliability of 0.85 and higher were obtained between air and bone thresholds for both methods of testing; that is, manual and computer-controlled audiometry. The same measurement between manual and computerized air-conduction pure-tone thresholds gave values ranging from 0.93 to 0.98. A principal-components analysis documented content validity of computerized audiometry. These results are interpreted as clear evidence of both reliability and validity of the BOBCAT procedure, with one exception: 6.7% of all observations are showing air-conduction computerized thresholds of -10 dB at one or more frequencies, with no confirmation by manual audiometry. This is attributed to insufficient randomization of intervals between stimuli, a problem which should be taken care of by new versions of the software.