Neonatal hearing screening using a smartphone-based otoacoustic emission device : a comparative study

Abstract

Early identification and intervention for childhood hearing loss assists in the reduction of delays in speech and language development. Otoacoustic emissions (OAEs) are commonly used for identifying hearing loss through newborn hearing screening (NHS). Recent technological advances in NHS have created more feasible and accessible services, however no objective smartphone-based screening applications (apps) have been validated. This study aimed to compare the screening outcomes of a smartphone-based OAE screening device to a commercially available OAE screening device. More specifically, the within-participant outcomes of the OAEs, in terms of screening concordance, signal, noise, and SNRs were compared to measure equivalence between the two devices. NHS at two tertiary public healthcare hospitals was conducted over a period of 8 months. The primary investigator followed a one-step screening protocol by means of which only OAEs were performed on each infant rather than a two-step screening protocol of OAEs and automated auditory brainstem responses (AABR). All infants less than three months of age were eligible for the study. DPOAE and TEOAE screenings were performed using the Otodynamics ILO v6 device (comparator) and the hearX OAE device. The screening technology namely the hearOAE and Otodynamics ILO, was used in an alternating manner on each day of the data collection period. Every participant was required to undergo OAE screening, which included either Distortion Product Otoacoustic Emission (DPOAE) screening or Transient Evoked Otoacoustic Emission (TEOAE) screening. A total of 176 infants (352 ears) (48.9% female) underwent TEOAE and DPOAE newborn hearing screening by a dedicated screener. The mean age at the time of screening was 4.5 days (SD 11.3). Statistically significant within-participant differences for DPOAE were measured for infants in the Neonatal Intensive Care Unit (NICU) on the pass/refer outcomes where infants who spent time in the NICU were 3.09 times more likely to refer DPOAE screening using the hearOAE device (p=0.029). Inter-device DPOAE comparison indicated no statistically significant difference in the refer rate between the devices (p=0.238). Similarly, inter-device differences for TEOAEs were measured for pass/refer outcomes. A statistically higher NHS pass rate was measured for TEOAEs with the hearOAE compared to the Otodynamics ILOs (p=0.009). The inter-device, within-participant diagnostic concordance was 89.7% and 85.0% for DPOAE and TEOAE respectively. The current study concluded that the hearOAE yielded outcomes comparable to the Otodynamics ILO v6 in terms of the overall pass and/or refer outcome. This verifies the performance of the novel smartphone-based OAE device and may facilitate increased accessibility of NHS services as mHealth is becoming a recognised alternative method in NHS programs globally.