The auditory brainstem response (ABR): an objective or subjective measure?

Abstract

Objectives: This thesis describes three studies investigating the consistency and reliability of ABR testing procedures, the extent of inter-and intra-rater variability and performance, the impact of additional case information on ABR testing results. Methods: The participants for all studies were audiologists who conduct ABR testing. Study 1 consisted of a 69-item survey regarding ABR testing protocols that was completed online. In studies 2 and 3, a test set of infant ABR traces was obtained from a large public peadiatric audiology clinic in Sydney, Australia. In Study 2, 61 audiologists estimated hearing threshold for 15 ABR cases that were each presented twice in order to estimate inter and intra-reader variability. In study 3, 14 audiologists were asked to estimate the hearing threshold for 16 infants twice in two sessions at least 5 months apart; with and without provision of standard clinical information. Results: Audiologists vary in how they conduct ABR testing and there are differences in the training period audiologists undertake before starting ABR testing. Audiologists show good levels of agreement when assessing repeated ABR traces, however, optimum performance was not achieved. No significant differences were found for sensitivity, specificity or accuracy when clinical history information was provided compared to when it was not provided. Conclusions: There is a need for greater emphasis on the importance of following evidence-based guidelines for ABR testing whenever it is possible. Audiologists’ experience and ABR training periods significantly predicts accuracy of ABR reading. Good levels of agreement were found between and within audiologists. ABR traces are interpreted with the same accuracy regardless of whether patients ‘history information is available or not. The data provided should contribute to an improvement in the service of infants hearing diagnostics and help reduce audiologists’ variability.