The aim of this study was to develop a Digit Triplet Test (DTT) using NZ English. The DTT is a hearing screening tool that uses spoken numbers presented in background noise to estimate speech recognition thresholds (SRTn). The NZ DTT will be made available via telephone or the internet, and will provide each person who completes the screening test with information about whether they should seek a professional hearing assessment. Normal-hearing participants (22 listeners) with hearing thresholds ≤20 dB HL were tested to establish the intelligibility of the individual digits at various signal-to-noise ratios (-20; -17.5; -15.0; -12.5; -10.0; -7.5; and -5.0 dB). The mid-points of the resulting psychometric functions were then used to adjust the level of each digit to achieve the same intelligibility. A SRT of -10.40 ± 1.75 dB SNR for the broadband presentation was established for the separate ear triplet test with the average slope of 17.3%/dB ± 3.9 %/dB for the ten test lists generated. The binaural ear DTT results were compared to best ear threshold PTA and found to have a highly significant correlation (r = 0.816, p<0.001) and a significant correlation to the QuickSIN sentence-in-noise test (r = 0.668, p<0.001). The binaural triplet test was found to have a sensitivity of 100% and specificity of 85%. The separate ear DTT results were compared to the best ear threshold pure tone audiometry and found to have a highly significant correlation (r = 0.809, p<0.001). The separate ear triplet test was found to have a sensitivity of 88% and specificity of 81% (1 – specificity = 0.187). The internet version of the DTT hearing screening test will provide New Zealanders with an easily accessible and objective test that will raise awareness about hearing and hopefully reduce the length of time people take before seeking advice about their hearing.

Hearing loss has a prevalence of 10.3% in New Zealand, with the Māori population being disproportionately affected compared to the non-Māori population. Hearing loss is an impairment that is under-recognised, under-reported and under-treated. This can be explained by the many existing barriers – the shortage of audiological services, financial cost to an individual seeking treatment, the stigma of both hearing loss and hearing aids, and healthcare seeking rates, particularly among the Māori population. This study aimed to develop a Māori language adaptive digit triplet test that could be offered remotely via the telephone and Internet as a hearing-screening test. Three sets of recordings were made of digit triplets spoken in te reo Māori by a female speaker. Two of these sets were selected for normalisation in speech noise. Normal-hearing participants (8 listeners) with hearing thresholds ≤20 dB HL were tested to establish the intelligibility of the individual recorded digits at various signal-to-noise ratios (-13, -10.5, -8 and -5.5 dB). Psychometric functions were fitted to the intelligibility data, and the digits in each position of the triplet that had the steepest slope were selected as the final test stimuli. The level of each selected digit was then adjusted to achieve equal intelligibility as measured at the midpoints of the psychometric functions. These digits were then assembled into eight equivalent lists of similar difficulty, ready for pilot testing. Due to low participant numbers, the pilot testing phase was not completed. Further development of this test continues as the focus of a follow-on study.

Presbycusis, or age-related hearing loss, is one of the most common chronic conditions to affect adults. On average individuals wait seven years from the time they notice a hearing impairment to the time they seek help from a hearing professional. This delay may have wide reaching implications for public health in the coming decades, as aging populations become more prevalent and as further research assesses the relationship between hearing loss and mental health conditions such as depression and dementia. The development of the New Zealand Hearing Screening Test (NZHST) aims to fulfil a need for a robust hearing screening test that individuals can access from home. This digit triplet test (DTT) will be particularly valuable for those in rural areas where audiological services are sparse and for those who have mobility issues which restrict attendance at clinical appointments. In order to accommodate as many New Zealanders as possible, the NZHST will have two versions, an internet version and a land-line telephone version; both of which can be delivered into their home in either New Zealand English or Te Reo Māori. This research is the third instalment in the development of the NZHST. The current research is divided into three parts; the verification of the New Zealand English DTT for the internet version, the pilot study for the Te Reo Māori DTT for the internet version, and the normalisation of the New Zealand English DTT for the telephone version. In the verification process, 50 individuals with various audiometric thresholds listened to 3 lists of 27 New Zealand English digit triplets, presented in three conditions; binaurally and to each ear separately via an internet interface. In the pilot study, 27 participants with various audiometric thresholds listened to 3 lists of 27 Te Reo Māori digit triplets via a software interface on a laptop computer. The normalisation process involved 10 individuals with normal hearing (average air-conduction pure tone thresholds of ≤ 20 dB HL) listening to 168 New Zealand English digit triplets under two different noise conditions; one as continuous speech noise and the other a noise with spectral and temporal gaps (STG noise) presented via a software interface on a laptop computer. Four conditions of the 168 digits were presented; once to each ear for the continuous noise, and once to each ear for the STG noise. Significant correlations were found between the binaural DTT and PTA (R = 0.66), and between the monaural ear DTT and PTA (R = 0.73) for the verification. The binaural DTT had a test sensitivity of 94% and a specificity of 88%. Pilot study correlation between binaural DTT and PTA was R = 0.61, and was R = 0.63 between monaural DTT and PTA; while the binaural sensitivity (100%) and specificity (100%) of the Te Reo DTT was affected by the very small number of participants with hearing loss (n = 4). The normalisation revealed that detection of the digit triplets was easier when STG noise (Lmid = -11.5 dB SNR, SD = 1.6 dB) was used as a masker, rather than continuous noise (Lmid = -8.9 dB SNR, SD = 1.4 dB).