Automated smartphone-based threshold audiometry has the potential to provide affordable audiometric services in underserved contexts where adequate resources and infrastructure are lacking. This study investigated the validity of the threshold version (hearTest) of the hearScreen™ smartphone-based application using inexpensive smartphones (Android OS) and calibrated supra-aural headphones. A repeated-measures, within-subject, study design was employed, comparing automated smartphone audiometry air conduction thresholds (0.5 to 8 kHz) to conventional audiometry thresholds. A total of 95 participants, with varying degrees of hearing sensitivity, were included in the study. 30 participants were adults, with known bilateral hearing losses of varying degrees (mean age of 59 years, 21.8 SD; 56.7% female). 65 participants were adolescents (mean age of 16.5 years, 1.2 SD; 70.8% female), of which 61 had normal hearing and 4 had mild hearing losses. Within the adult sample, 70.6% of thresholds obtained through smartphone and conventional audiometry corresponded within 5 dB. There was no significant difference between smartphone (6.75 min average, 1.5 SD) and conventional audiometry test duration (6.65 min average, 2.5 SD). Within the adolescent sample, 84.7% of audiometry thresholds obtained at 0.5, 2 and 4 kHz corresponded within 5 dB. At 1 kHz 79.3% of the thresholds differed by 10 dB or less. There was a significant difference (p&#060.01) between smartphone (7.09 min, 1.2 SD) and conventional audiometry test duration (3.23 min, 0.6 SD). The hearTest application using calibrated supra-aural headphones provided valid air conduction hearing thresholds. Therefore, it is evident that using inexpensive smartphones with calibrated headphones provides a cost-effective way to provide access to threshold air conduction audiometry. / Dissertation (M Communication Pathology)--University of Pretoria, 2016. / Speech-Language Pathology and Audiology / M Communication Pathology / Unrestricted

Decentralised detection and monitoring of hearing loss can be supported by new mHealth technologies using automated testing, which can be facilitated by minimally trained persons. These technologies may prove particularly useful in an infectious disease (ID) clinic setting where patients are at high risk for hearing loss. The current study aimed to evaluate the clinical utility of mobile and automated audiometry hearing health technology in an ID clinic setting. The current study was exploratory as it aimed to determine whether smartphone automated audiometry and South African English Digits-In-Noise (SA Eng DIN) smartphone applications could be utilised in an infectious disease clinic setting to monitor an HIV-related hearing loss in a feasible and time efficient way. Smartphone automated audiometry (hearTest™) and speech-in-noise testing (SA English Digits-In-Noise (DIN) test) were compared with manual audiometry at 2, 4, and 8 kHz. Smartphone automated audiometry and the DIN test were repeated to determine the test re-test reliability. Two hundred subjects (73% female and 27% male) were enrolled. Fifty participants were re-tested with the smartphone applications. Participants’ ages ranged from 18 to 55 years with a mean age of 44.4 (8.7 SD). Threshold comparisons were made between smartphone audiometry testing and manual audiometry. Smartphone automated audiometry, manual audiometry, and test re-test measures were compared to determine the statistical significance of any differences observed using the Wilcoxon signed-ranked test. Spearman rank correlation test was used to determine the relationship between the smartphone applications and manual audiometry, as well as for test re-test measurements. For all participants, 88.2% of thresholds corresponded within 10 dB or less between smartphone audiometry and manual audiometry. There was a significant difference (p>0.05) between smartphone and manual audiometry for the right ear at 4 and 8 kHz and the left ear at 2 and 4 kHz respectively. No significant difference was noted (p>0.05) between test and re-test measures of smartphone technology except at 4kHz in the right ear in smartphone automated audiometry. The absolute average difference between the initial and re-test of DIN testing was 1.2 dB (1.5 SD). No significant difference was noted in the test re-test measures of the DIN test (p < vii 0.05). A correlation coefficient of 0.56 was present in the DIN test re-test measures when the Spearman rank correlation test was administered. Smartphone audiometry with calibrated headphones provides reliable results and can be used as a baseline and monitoring tool at ID clinics. / Dissertation (MA)--University of Pretoria, 2017. / Speech-Language Pathology and Audiology / MA / Unrestricted

Background: Disabling hearing loss is a global burden. This burden is worsened by the emergence of multi-drug resistant tuberculosis (MDR-TB). Some of the medications used to treat MDR-TB are damaging to the cochlea and auditory nerve (ototoxic) and can lead to permanent hearing loss and/or balance disorders. Ototoxicity monitoring aims to reduce this burden by preventing or minimising the damage caused by ototoxic treatment as it can progress and worsen speech perception difficulties. However, the proposed test battery for ototoxicity monitoring is lengthy and demands active participation which is not ideal for ill patients (such as those on MDR-TB treatment). The Sensitive Range of Ototoxicity (SRO) technique is recommended to shorten the test time. The SRO consists of seven consecutive relatively high frequencies determined from the highest frequency the participant responded to. The SRO technique is time efficient. Although the SRO technique provides the prospect of a shortened test battery, there is still a global lack of audiologists. Automated audiometry is a vital application for testing especially when audiologists are not available to physically do the test. Automated audiometry has been previously validated. Clinically, automated audiometry is objective and allows for standardisation. Even though automated audiometry helps improve access to monitoring more patients, patient preference is an important factor when using automated audiometry to ensure patient-centred care is not compromised. Aims and Objectives: This study aimed to investigate the specificity and sensitivity of the SRO technique with automated audiometry compared to the gold standard (manual audiometry). This comparison was made by firstly, determining the testing time efficiency and the correlation of thresholds obtained with the different test methods and, secondly, testing the diagnostic value of automated audiometry using the SRO technique. The incidence of an ototoxicity-induced hearing loss was described by determining the time interval between starting ototoxic MDR-TB treatment and the onset of a significant threshold shift (STS) according to ASHA's criteria. Lastly, the test method preference of the participants with MDR-TB was described and compared using a short exit survey. Study Design: A prospective repeated-measures study design was used. Participants were chosen based on a risk factor (i.e. exposure to ototoxic medication) for an outcome of interest (i.e. the presence or absence of an STS). With a repeated measures study, multiple tests using different test methods can be compared with the same sample. Participants: Twenty-seven in-patients at Brooklyn Chest Hospital and DP Marais TB Hospital with normal hearing and on MDR-TB medication were included in the study. Their age range was from 19 to 51 years old with an average age of 33 years old. Non-probability convenience sampling was used as it was cost-effective, reduced data collection time and was relatively easy to execute. Data collection materials and procedures: The procedure for data collection included weekly follow-up testing for a maximum of four weeks. The test battery was as follows: an auditory symptom questionnaire, otoscopy examination, and manual and automated audiometry using the SRO technique with a fifteen-minute break in between. Participants were tested with the KUDUwave ™ in a non-sound treated room. The frequency range was determined with the SRO technique. If an STS was obtained, the patient was discharged from the study after completing an exit survey. Statistics: Analysis included descriptive statistics and inferential statistics. A Bonferroni corrected p-value (initially p ≤ 0.05) was used. Manual and automated audiometry thresholds were compared using the Pearson's Correlation Coefficient test. Manual and automated audiometry testing time and threshold means were compared using paired sample's t-tests. The diagnostic value of automated audiometry with the SRO technique was assessed with Receiver Operating Characteristics (ROC) Curves. Results: Manual audiometry was statistically more time-efficient compared to automated audiometry by an average of one minute and ten seconds (t (94) = -5.44; p< 0.003). There was a strong positive correlation for both left and right ears between the thresholds' obtained from manual and automated audiometry at 8kHz to 16 kHz (df> 28 = r > 0.70, p< 0.003). Automated audiometry was found to be a fair diagnostic test (area under the curve was 0.75; p= 0.002). Also, the ROC curve revealed that automated audiometry had a sensitivity of 61% and specificity of 90% when compared to manual audiometry (gold standard). Only participants that started data collection within 31 days after starting their MDR-TB treatment were included in the analysis of determining the incidence of an ototoxicity-induced hearing loss (n= 24 ears). This study found that 41.67% of ears (n= 10) had an ototoxicity-induced hearing loss. A box and whisker plot revealed that data was skewed to the right (i.e. more variation in data between the median and the maximum values) and that the median number of days for an ototoxicity-induced hearing loss to appear was 33 days. Secondly, 55.55% of participants (n=15 out of 27) reported auditory symptoms before data collection commencement. Aural fullness was the most reported symptom (n= eight out of 15). Ten out of 15 (66.66%) participants that reported auditory symptoms obtained an ototoxicity-induced hearing loss. Lastly, most participants (i.e. 13 out of 19; 68.42%) that completed the exit survey had no preference between manual or automated audiometry. The common rationale among these participants was “No difference noted.” Conclusion: This research study has revealed that manual audiometry was more time-efficient compared to automated audiometry in patients with MDR-TB. Also, automated audiometry was a fair diagnostic test. It may aid in reducing the disproportionate audiologist to patient ratio, especially in a developing country. However, manual audiometry (with the SRO technique) is more clinically appropriate in patients that are difficult-to-test. Secondly, audiometric settings can be changed to accommodate testing frequencies in 1/6 octaves so that the SRO technique can be clinically adopted. An ototoxicity-induced hearing loss seems to appear 33 days after ototoxic MDR-TB treatment commencement. Aural fullness was a commonly reported symptom among participants with MDRTB. Aural fullness is omnipresent in peripheral auditory pathologies. Therefore, auditory symptoms reported by patients' needs a comprehensive audiological investigation. Lastly, more research is needed on how patients (and clinicians) experience the advances in technology innovation especially in audiology where technology innovation is continuously evolving.