# An Initial Validation of Community-Based Air-Conduction Audiometry in Adults With Simulated Hearing Impairment Using a New Web App, DigiBel: Validation Study

## TL;DR

A new web app called DigiBel was tested for its ability to detect hearing impairment in adults using simulated hearing loss, showing promising results for community-based screening.

## Contribution

The study introduces and validates a novel web-based audiometry tool, DigiBel, for detecting hearing impairment in community settings.

## Key findings

- DigiBel demonstrated 100% sensitivity and 72.73% specificity in detecting hearing impairment of more than 20 dB.
- Threshold measurements were mostly accurate, with only minor overestimation at 4000 and 8000 Hz.
- Participants rated the DigiBel test as good or excellent and expressed confidence in using it at home.

## Abstract

Approximately 80% of primary school children in the United States and Europe experience glue ear, which may impair hearing at a critical time for speech acquisition and social development. A web-based app, DigiBel, has been developed primarily to identify individuals with conductive hearing impairment who may benefit from the temporary use of bone-conduction assistive technology in the community.

This preliminary study aims to determine the screening accuracy and usability of DigiBel self-assessed air-conduction (AC) pure tone audiometry in adult volunteers with simulated hearing impairment prior to formal clinical validation.

Healthy adults, each with 1 ear plugged, underwent automated AC pure tone audiometry (reference test) and DigiBel audiometry in quiet community settings. Threshold measurements were compared across 6 tone frequencies and DigiBel test-retest reliability was calculated. The accuracy of DigiBel for detecting more than 20 dB of hearing impairment was assessed. A total of 30 adults (30 unplugged ears and 30 plugged ears) completed both audiometry tests.

DigiBel had 100% sensitivity (95% CI 87.23-100) and 72.73% (95% CI 54.48-86.70) specificity in detecting hearing impairment. Threshold mean bias was insignificant except at 4000 and 8000 Hz where a small but significant overestimation of threshold measurement was identified. All 24 participants completing feedback rated the DigiBel test as good or excellent and 21 (88%) participants agreed or strongly agreed that they would be able to do the test at home without help.

This study supports the potential use of DigiBel as a screening tool for hearing impairment. The findings will be used to improve the software further prior to undertaking a formal clinical trial of AC and bone-conduction audiometry in individuals with suspected conductive hearing impairment.