PreprintPDF Available
Preprints and early-stage research may not have been peer reviewed yet.

Abstract

Objective: The digits-in-noise test (DIN) has become increasingly popular as a consumer-based method to screen for hearing loss. Current versions of all DINs either test ears monaurally or present identical stimuli binaurally (i.e., diotic noise and speech, NoSo). Unfortunately, presentation of identical stimuli to each ear inhibits detection of unilateral sensorineural hearing loss (SNHL), and neither diotic nor monaural presentation sensitively detects conductive hearing loss (CHL). Following an earlier finding of enhanced sensitivity in normally hearing listeners, this study tested the hypothesis that interaural antiphasic digit presentation (NoSπ) would improve sensitivity to hearing loss caused by unilateral or asymmetric SNHL, symmetric SNHL, or CHL. Design: This cross-sectional study, recruited adults (18-84 years) with various levels of hearing, based on a four-frequency pure tone average (PTA) at 0.5, 1, 2 and 4kHz. The study sample was comprised of listeners with normal hearing (n=41; PTA ≤ 25 dB HL in both ears), symmetric SNHL (n=57; PTA > 25 dB HL), unilateral or asymmetric SNHL (n=24; PTA > 25 dB HL in the poorer ear) and CHL (n=23; PTA > 25 dB HL and PTA air-bone gap ≥ 20 dB HL in the poorer ear). Antiphasic and diotic speech reception thresholds (SRTs) were compared using a repeated-measures design. Results: Antiphasic DIN was significantly more sensitive to all three forms of hearing loss than the diotic DIN. SRT test-retest reliability was high for all tests (ICC r > 0.89). Area under the receiver operating characteristics (ROC) curve for detection of hearing loss (> 25 dB HL) was higher for antiphasic DIN (0.94) than for diotic DIN (0.77) presentation. After correcting for age, PTA of listeners with normal hearing or symmetric SNHL was more strongly correlated with antiphasic (rpartial[96]=0.69) than diotic (rpartial=0.54) SRTs. Slope of fitted regression lines predicting SRT from PTA was significantly steeper for antiphasic than diotic DIN. For listeners with normal hearing or CHL, antiphasic SRTs were more strongly correlated with PTA (rpartial[62]=0.92) than diotic SRTs (rpartial[62]=0.64). Slope of regression line with PTA was also significantly steeper for antiphasic than diotic DIN. Severity of asymmetric hearing loss (poorer ear PTA) was unrelated to SRT. No effect of self-reported English competence on either antiphasic or diotic DIN among the mixed first-language participants was observed. Conclusions: Antiphasic digit presentation markedly improved the sensitivity of the DIN test to detect SNHL, either symmetric or asymmetric, while keeping test duration to a minimum by testing binaurally. In addition, the antiphasic DIN was able to detect CHL, a shortcoming of previous monaural or binaurally diotic DIN versions. The antiphasic DIN is thus a powerful tool for population-based screening. This enhanced functionality combined with smartphone delivery could make the antiphasic DIN suitable as a primary screen that is accessible to a large global audience.
1
Improving Sensitivity of the Digits-in-Noise Test using
Antiphasic Stimuli
Karina C. De Sousa,1 De Wet Swanepoel,1,2 David R. Moore3,4, Hermanus
Carel Myburgh5 and Cas Smits6
1 Department of Speech-Language Pathology and Audiology, University
of Pretoria, Pretoria, Gauteng, South Africa.
2 Ear Sciences Centre, School of Surgery, University of Western
Australia, Nedlands, Australia.
3 Communication Sciences Research Center, Cincinnati Childrens’
Hospital Medical Center and University of Cincinnati, Ohio, USA.
4 Manchester Centre for Audiology and Deafness, University of
Manchester, United Kingdom.
5 Department of Electrical, Electronic and Computer Engineering,
University of Pretoria, Pretoria, Gauteng, South Africa.
6 Amsterdam UMC, Vrije Universiteit Amsterdam, Otolaryngology-Head
and Neck Surgery, Ear and Hearing, Amsterdam Public Health
research institute, De Boelelaan 1117, Amsterdam.
Financial Disclosures/ Conflict of Interest:
This research was funded by the National Institute of Deafness and Communication
Disorders of the National Institutes of Health under award number 5R21DC016241-
02. The 2nd author’s relationship with the hearX Group and hearZA includes equity,
consulting, and potential royalties. The third author’s relationship with the hearX
Group and hearZA includes equity, consulting and potential royalties.
Address correspondence to Amsterdam UMC, Vrije Universiteit Amsterdam,
Otolaryngology-Head and Neck Surgery, Ear and Hearing, Amsterdam Public Health
research institute, De Boelelaan 1117, Amsterdam. E-mail: C.Smits@vumc.nl
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
2
ABSTRACT
Objective: The digits-in-noise test (DIN) has become increasingly popular as a
consumer-based method to screen for hearing loss. Current versions of all DINs either
test ears monaurally or present identical stimuli binaurally (i.e., diotic noise and
speech, NoSo). Unfortunately, presentation of identical stimuli to each ear inhibits
detection of unilateral sensorineural hearing loss (SNHL), and neither diotic nor
monaural presentation sensitively detects conductive hearing loss (CHL). Following
an earlier finding of enhanced sensitivity in normally hearing listeners, this study tested
the hypothesis that interaural antiphasic digit presentation (NoSπ) would improve
sensitivity to hearing loss caused by unilateral or asymmetric SNHL, symmetric SNHL,
or CHL.
Design: This cross-sectional study, recruited adults (18-84 years) with various levels
of hearing, based on a four-frequency pure tone average (PTA) at 0.5, 1, 2 and 4kHz.
The study sample was comprised of listeners with normal hearing (n=41; PTA ≤ 25 dB
HL in both ears), symmetric SNHL (n=57; PTA > 25 dB HL), unilateral or asymmetric
SNHL (n=24; PTA > 25 dB HL in the poorer ear) and CHL (n=23; PTA > 25 dB HL and
PTA air-bone gap ≥ 20 dB HL in the poorer ear). Antiphasic and diotic speech
reception thresholds (SRTs) were compared using a repeated-measures design.
Results: Antiphasic DIN was significantly more sensitive to all three forms of hearing
loss than the diotic DIN. SRT test-retest reliability was high for all tests (ICC r > 0.89).
Area under the receiver operating characteristics (ROC) curve for detection of hearing
loss (> 25 dB HL) was higher for antiphasic DIN (0.94) than for diotic DIN (0.77)
presentation. After correcting for age, PTA of listeners with normal hearing or
symmetric SNHL was more strongly correlated with antiphasic (rpartial[96]=0.69) than
diotic (rpartial=0.54) SRTs. Slope of fitted regression lines predicting SRT from PTA
was significantly steeper for antiphasic than diotic DIN. For listeners with normal
hearing or CHL, antiphasic SRTs were more strongly correlated with PTA
(rpartial[62]=0.92) than diotic SRTs (rpartial[62]=0.64). Slope of regression line with PTA
was also significantly steeper for antiphasic than diotic DIN. Severity of asymmetric
hearing loss (poorer ear PTA) was unrelated to SRT. No effect of self-reported English
competence on either antiphasic or diotic DIN among the mixed first-language
participants was observed
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
3
Conclusions: Antiphasic digit presentation markedly improved the sensitivity of the
DIN test to detect SNHL, either symmetric or asymmetric, while keeping test duration
to a minimum by testing binaurally. In addition, the antiphasic DIN was able to detect
CHL, a shortcoming of previous monaural or binaurally diotic DIN versions. The
antiphasic DIN is thus a powerful tool for population-based screening. This enhanced
functionality combined with smartphone delivery could make the antiphasic DIN
suitable as a primary screen that is accessible to a large global audience.
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
4
INTRODUCTION
Hearing loss presents a significant global health burden as the 4th leading contributor
to years lived with disability (Vos et al. 2016). Mounting evidence demonstrates
significant associations between hearing loss, depression (Fellinger et al. 2012),
unemployment (Ruben 2015), risk for hospitalization (Genther et al. 2013; Reed et al.
2018) and cognitive decline and dementia (Lin et al. 2011; Livingston et al. 2017).
Early detection is an essential first step to ameliorate the functional impairment of
hearing loss, yet a high proportion of cases remains undetected and untreated (Ki-
Moon 2016; Mackenzie and Smith 2009). Contributing to the disparity is lack of routine
adult hearing screening programs and rehabilitation options that are either unavailable
or prohibitively expensive (Chou et al. 2011; Wilson et al. 2017).
Poor awareness of hearing loss and existing models of clinic-based adult screening
among the lay public also contribute to hearing healthcare inaccessibility (Lin et al.
2016). In efforts to increase and decentralize access to detection of hearing loss,
screening methods such as the digits-in-noise test (DIN), as an internet or landline
phone-based hearing screen have been employed (Smits et al. 2004; Jansen et al.
2010; Watson et al. 2012; Zokoll et al. 2012). The DIN is a speech-in-noise test that
uses digit triplets (e.g. 5-9-2), typically presented in steady speech-shaped noise, to
measure the speech reception threshold (SRT), expressed in dB signal-to-noise ratio
(dB SNR), where a listener can recognize 50% of the digit triplets correctly. Compared
to pure tone audiometry or speech recognition in quiet, speech recognition in noise
has the advantage of being more characteristic of a person’s hearing ability in real-life
situations (Grant et al. 2013). Furthermore, DIN assessment of sensorineural hearing
loss (SNHL) correlates highly with pure tone audiometry and eliminates the need for
a soundproof booth, calibrated equipment and a test administrator (Smits et al. 2004;
Jansen et al. 2010; Potgieter et al. 2016, 2018; Koole et al. 2016).
The DIN was first developed as a national landline telephone test in the Netherlands
(Smits et al. 2004) and later also implemented as an internet-based test (Smits et al.
2006). Highly correlated with the audiometric pure tone average (r=0.77) it
demonstrated sensitivity and specificity of more than 90% to detect sensorineural
hearing loss (Smits et al. 2004). Four months after its release, the DIN saw
considerable uptake with more than 65,000 tests taken (Smits et al. 2005),
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
5
demonstrating its role and potential as a large-scale hearing screening tool available
to the public. Using simple digits, the test does not require a high degree of linguistic
competence (Kaandorp et al. 2016). Various language versions of the DIN have been
developed, including British-English (Hall, 2006), American-English (Watson et al.
2012); Polish (Ozimek et al. 2009), French (Jansen et al. 2010) and German (Zokoll
et al. 2011).
Despite the success of the DIN in several countries, the need of landline telephones
to conduct testing can be problematic, especially in low-and-middle income countries
like South Africa where landline penetration is poor (STATSSA 2013). On the other
hand, global access to smartphones by adults is estimated to be 80% by the year
2020, providing a modern-day alternative (The Economist 2015). Whereas mobile
phone penetration is much higher, the cost to complete the test via a mobile phone
call could be more expensive. An alternative is to offer the DIN as a downloadable
smartphone application, allowing access to high fidelity broadband signals as opposed
to bandwidth signals used in standard telephone networks (Potgieter et al. 2016), and
removing the need for cellular connectivity once uploaded. While applicable
worldwide, using a mobile platform could potentially address the mostly nonexistent
access to hearing screening in low-and-middle income countries. In sub-Saharan
Africa, for instance, there is only one audiologist for every million people (Mulwafu et
al. 2017). As a result, the South African English DIN was developed and released as
the national hearing screening application in 2016, downloadable on iOS and Android
smartphones, called hearZATM (Potgieter et al. 2016; De Sousa et al. 2018). This
binaural test version allows for testing under 3 minutes, with high sensitivity (> 80%)
to detect SNHL (Potgieter et al. 2016; Potgieter et al. 2018).
There has been a growing interest in increasing the efficiency and sensitivity of
existing DINs using various test modifications. Using a fixed-SNR procedure, Smits
(2017) showed that the number of digit triplets in a DIN could be reduced to as few as
8 trials, without compromising sensitivity and specificity but sacrificing accurate
estimation of the SRT. Furthermore, with the early appearance and high prevalence
of high frequency hearing loss, use of low-pass filtered masking noise to improve
sensitivity of the DIN to high frequency hearing loss has been investigated, showing
either higher (Vlaming et al. 2014) or similar (Vercammen et al. 2018) area under the
receiver operating characteristics (ROC) curve compared to DINs with standard
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
6
speech-shaped noise. Therefore, when using homogenized digits to ensure high test-
retest reliability, these modifications could make the DIN test more applicable to
persons with noise-induced or age-related hearing loss (Vlaming et al. 2014).
Current versions of all DINs either sequentially test each ear (monaurally) or present
the test stimuli binaurally and identically to each ear (homophasic or diotic). This
binaural DIN setup allows for rapid testing in approximately 3 minutes, whereas
sequential testing of each ear doubles test time and may thus reduce uptake and
completion. Using diotic presentation may, however, preclude detection of unilateral
or asymmetric sensorineural hearing loss (SNHL). These listeners may pass the diotic
DIN test because performance is largely based on the functionally better ear (Potgieter
et al. 2018). Furthermore, both monaural and diotic testing is insensitive to the
attenuation caused by conductive hearing loss (CHL) because most DINs are
presented at suprathreshold intensities. To improve the sensitivity of the DIN,
especially for listeners with unilateral, asymmetrical SNHL and CHL, this study
evaluated the use of a DIN test paradigm using digits that are phase inverted
(antiphasic) between the ears, while leaving the masking noise interaurally in-phase.
Such a configuration of stimuli (NoSπ) was shown to improve DIN SRTs in normal
hearing listeners (Smits et al. 2016).
Sensitivity differences between diotic and antiphasic auditory stimulus presentations
are commonly known as the binaural masking level difference (Hirsh 1948). Before
the widespread use of the auditory brainstem response, binaural masking level
difference was employed to distinguish between different types of hearing loss (Olsen
et al. 1976; Wilson et al. 2003). Binaural masking level difference was reported to be
poorer for listeners with various types and configurations of hearing loss compared to
normal hearing controls. Wilson and colleagues (1985) investigated speech masking
level difference for people with unilateral SNHL. In the diotic condition (NoSo), only
slight SNR variations were observed across a range of interaural level differences.
However, in the antiphasic condition (NoSπ), SNRs became worse with increasing
interaural level differences.
Smits and colleagues (2016) examined SRTs in different listening conditions for the
Dutch and American English DIN among normal hearing listeners. Results indicated
that the threshold advantage over monotic presentation provided by diotic (NoSo)
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
7
presentation was small ( 1 dB). However, the use of antiphasic digits (NoSπ) provided
a further 5 dB advantage. Listeners with unilateral SNHL or CHL are not expected
to have full access to the antiphasic advantage due to subtle timing irregularities
caused by peripheral hearing loss, either sensorineural (Jerger et al. 1984; Thornton
et al. 2012; Wilson et al. 1985) or conductive (Hartley and Moore 2003; Jerger et al.
1984). In cases of symmetric hearing loss, the antiphasic advantage is expected to
decrease as the degree of hearing loss increases because of increasing threshold and
timing cue deterioration (Wilson et al. 1994). These findings support the idea that
antiphasic digit presentation could sensitize the DIN for a wider range of hearing loss
types while using a single binaural test. This would improve the function of current
consumer-based DINs.
The objective of this study was, therefore, to determine whether antiphasic digit
presentation improves detection of hearing loss relative to the diotic presentation.
MATERIALS AND METHODS
Study design and participants
A cross-sectional, repeated-measure study of the DIN SRT comparing diotic and
antiphasic presentation within and between listeners of varying types and degrees of
hearing loss was conducted. Listeners were recruited from a student population, a
University clinic, and hospital and private practices in the Gauteng province of South
Africa. Adults (18-84 years; Table 1) with various levels of hearing were recruited,
based on a four frequency (0.5,1,2 and 4kHz) pure tone average (PTA). The study
sample included normal hearing (n=41; pure tone average (PTA) 25 dB HL in both
ears), symmetric SNHL (n=57; PTA > 25 dB HL) and unilateral or asymmetric SNHL
(n=24; PTA > 25 dB HL in the poorer ear). The better ear PTA of listeners with
asymmetric SNHL did not exceed 45 dB HL. A sample of listeners with CHL (n=23;
PTA > 25 dB HL and PTA air-bone gap 20 dB HL in the poorer ear) was also
recruited, including 3 listeners with symmetric and 20 with unilateral or asymmetric
hearing loss. Bone conduction pure tone average thresholds (0.5,1,2,4 kHz) for the
poorer ear did not exceed 25 dB HL, except for one listener with CHL with poorer ear
bone conduction PTA of 28 dB HL. Asymmetric hearing loss was defined as an
interaural difference >10 dB (PTA). Hearing sensitivity categories were based on
poorer ear PTA and categorized as excellent (0-15 dB HL), minimal (16-25 dB HL),
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
8
mild (26-40 dB HL), moderate (41-55 dB HL) and severe-to-profound (56-120 dB HL).
For analyses, the ‘excellent’ and ‘minimal’ categories were combined into a single
‘normal’ category. Listeners had various levels of English-speaking competence. Non-
native English speakers self-reported their level of competence on a non-standardized
scale from 1-10, a higher score indicating better competence (Potgieter et al. 2018).
The Health Sciences Research Ethics Committee, University of Pretoria approved the
study protocol (number 58/2017). All eligible participants were informed on the study
aims and procedures and provided consent before participation.
Procedures and Equipment
The smartphone application for the South African English DIN was adapted for
antiphasic stimulus presentation. Original homogenized diotic digits were phase
reversed for antiphasic presentation. The phase inversion was completed in MatLab
by multiplying each sample in one channel of the digit triplet sound file by -1. The DIN
application was designed in Android Studio version 2.3.0 and written in Java version
1.8.0, consistent with the original hearZA App. The application stored a list of 120
different digit-triplets, randomly selected for presentation at the beginning of each test
(Potgieter et al. 2016). Randomized triplet selection was done with replacement,
meaning that the same triplet could occur more than once in one test. Triplets were
presented with 500 ms silent intervals at the beginning and end of each digit-triplet.
Successive digits were separated by 200 ms of silence with 100 ms of jitter (Potgieter
et al. 2016). The test used a fixed noise level and variable speech level when triplets
with negative SNRs were presented. To prevent clipping of the signal, the speech level
was fixed, and the noise level varied once the SNR became positive (Potgieter et al.
2016). The speech-weighted masking noise was delivered interaurally in-phase, and
the digits were either in-phase (diotic; NoSo) or were phase inverted between the two
ears (antiphasic; NoSπ). To prevent possible learning of the masking noise (Lyzenga
and Smits 2011), noise ‘freshness’ was ensured for each trial by creating a long noise
file and selecting successive fragments from a random offset within the first 5 seconds.
Both diotic and antiphasic versions of the DIN consisted of 23 digit-triplets. The SNR
varied in fixed step sizes (4 dB SNR for the first three steps, thereafter continuing in 2
dB steps) starting at 0 dB SNR using a one-up one-down staircase procedure, tracking
the SNR at which 50% of the digit triplets were correctly identified (Smits et al. 2004;
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
9
Potgieter et al. 2016). For the first three steps, SNR became progressively more
negative by 4 dB per step for correct responses but increased by 2 dB per step for
incorrect responses. A digit-triplet was only considered correct when all digits were
entered correctly. The SRT was calculated by averaging the last 19 SNRs, in line with
the currently used hearZA test.
After completion of pure tone audiometry, participants completed five DIN tests, each
lasting about 3 minutes, on a Samsung Trend Neo smartphone coupled with
manufacturer supplied (wired) earbuds in a quiet, office-like room. The first training
test used antiphasic presentation. The remaining four DIN tests alternated between
antiphasic and diotic DIN, with a test and retest for each participant. The test order
was therefore: (1) antiphasic training list, (2) antiphasic test, (3) diotic test, (4)
antiphasic re-test and (5) diotic re-test.
Statistical Analysis
Statistical analysis was done using the Statistical Package for the Social Sciences
(IBM SPSS v25.0). A sample size of 122 listeners (24 with normal hearing PTA ≤ 25
dB HL, 24 with asymmetric hearing loss and 74 with either symmetric normal hearing
PTA ≤ 25 dB HL or symmetric sensorineural hearing loss with PTA ≥ 26 dB HL) would
provide a medium effect size (Cohen’s f = 0.25), with 80% statistical power at two-
tailed significance level of 0.05, to test both hypotheses. The sample of 23 listeners
with CHL was subsequently added.
The effect of test condition (i.e., diotic or antiphasic) and hearing loss category (i.e.,
type and symmetry of hearing loss) on the SRT was assessed using repeated-
measures analysis of variance. Post hoc comparisons used Bonferroni adjustment for
multiple comparisons. In cases where sphericity was violated, Greenhouse-Geisser
corrections were applied. Analysis of covariance was used to determine effects of age
and English-speaking competence on the diotic and antiphasic SRT. General linear
regression was used to test whether the slope of the relation between PTA and SRT
differed between antiphasic and diotic testing. The effect of test repetition on
antiphasic SRT was investigated using a paired sample t-test. Intraclass correlation
coefficients (ICC) were calculated and were based on a mean rating of the number of
observations (i.e. test and retest; k=2) of both diotic and antiphasic test conditions,
absolute agreement, and a two-way mixed-effects model. In addition, measurement
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
10
error between test-retest for diotic and antiphasic presentation was calculated by
determining quadratic mean (2) of within-subject standard deviations for the test-
retest measures. All subsequent analyses were conducted by averaging the test and
retest SRT values for the diotic and antiphasic DIN. Associations between poorer ear
PTA and SRT were examined using Pearson’s partial correlation. Receiver operating
characteristics (ROC) curves were calculated to determine the sensitivity and
specificity of the DIN tests for different cutoff values, to detect mild hearing loss and
worse (PTA > 25 dB HL) and moderate hearing loss and worse (PTA > 40 dB HL).
SRT cut-off values corresponding to reasonably high sensitivity and specificity were
chosen, while demonstrating the trade-off between sensitivity and specificity (i.e.
higher sensitivity with consequent lower specificity).
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
11
RESULTS
Listeners with normal hearing had lower SRTs than those with hearing loss using both
diotic and antiphasic testing (Fig. 1). However, antiphasic testing was significantly
more sensitive to all three forms of hearing loss than diotic testing (Table 1).
Across all hearing categories, after controlling for age, poorer ear PTA was
significantly correlated to both diotic and antiphasic SRT (p < 0.001). The correlation
was, however, stronger for antiphasic (rpartial[145]=0.82] than diotic SRT
(rpartial[145]=0.44). For listeners with either normal hearing or symmetric SNHL, poorer
ear PTA was significantly (p < 0.001) correlated with both antiphasic (rpartial[96]=0.69)
and diotic (rpartial[96]=0.54) SRTs (Fig. 2). However, the slope of the fitted regression
was significantly steeper for antiphasic SRTs (t(1)=7.79.14, p < 0.001). Antiphasic
SRTs of listeners with normal hearing or CHL, were more strongly correlated to poorer
ear PTA (rpartial[62]=0.92) than diotic SRTs (rpartial[62]=0.54). The slope of the fitted
regression was also significantly steeper for antiphasic compared to diotic SRTs
(t[1]=11.84, p < 0.0001), indicative of greater sensitivity of the antiphasic DIN. The
severity of unilateral or asymmetric SNHL (poorer ear PTA) was unrelated to SRT. For
the diotic DIN, there was substantial overlap between the SRTs of normal hearing
listeners and those in each of the three hearing loss groups (Fig. 2A), even for PTAs
in the moderate or greater hearing loss ranges (Table 2). The SRT overlap was less
substantial for the antiphasic DIN, with listeners with mild poorer ear hearing loss
corresponding in SRTs to the normal hearing group (Fig. 2B).
ROC analysis, including poorer ears of all participants, (Fig. 3) showed higher areas
under the curve for antiphasic DIN compared to diotic DIN to detect PTA >25 dB HL
(0.95; 95% CI, 0.91 to 0.98 vs 0.78; 95% CI, 0.69 to 0.86) and >40 dB HL (0.96; 95%
CI, 0.93 to 0.99 vs 0.80; 95% CI, 0.73 to 0.87). Antiphasic DIN was, therefore, more
sensitive and specific to hearing loss (of either type and symmetry) compared to diotic
DIN. SRT cut-offs in Table 3 demonstrate the trade-off between sensitivity and
specificity to detect PTA > 25 dB HL and > 40 dB HL.
Antiphasic DIN test repetition produced a significant mean SRT improvement (0.9 dB
SNR; 95% CI 0.61 to 1.4) across hearing categories following the presentation of the
initial antiphasic training list (t[144]=5.1, p < 0.001; Fig. 1). However, between the
subsequent test and retest, the mean SRT difference was not significant (p = 0.86).
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
12
Similarly, diotic test and retest showed no significant SRT difference (p = 0.6). SRT
test-retest reliability was high for listeners with normal hearing or SNHL for both diotic
DIN (ICC=0.89; 95% CI 0.85 to 0.93) and antiphasic DIN (ICC=0.94; 95% CI, 0.91 to
0.96). Listeners with CHL had high test-retest reliability for antiphasic DIN, with ICC of
0.88 (95% CI, 0.72 to 0.95; p < 0.001), but had poorer ICC of 0.61 for homophasic DIN
(95% CI, 0.09 to 0.83; p < 0.05). Diotic DIN had lower measurement error (1.1 dB;
95% CI 0.9 to 1.2) than the antiphasic DIN (1.4 dB; 95% CI 1.2 to 1.5) for the whole
sample, but the variance between listeners was much higher for the antiphasic DIN
than for the diotic DIN (Table 2).
The effect of competence in the English language on SRT was assessed by dividing
listeners into high competence (>7; n=73) and lower competence (≤7; n=72) groups.
Controlling for poorer ear PTA and age, no significant SRT difference (p = 0.16) was
found between the two groups for either the diotic DIN (F[1,141]=2.47, partial η2=0.02)
or the antiphasic DIN (F[1,141]=1.98, partial η2 = 0.02).
Table 1. Analysis of variance statistics for the effect of test presentation and category
of hearing loss.
df
F
Sig
Test Type
(Diotic vs Antiphasic)
1, 141
497.06
< .001
0.78
Hearing Category
3,141
31.88
< .001
0.41
Test Type*Hearing
Category
3, 141
57.81
< .001
0.55
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
13
Table 2. Diotic and antiphasic DIN SRT for listeners with normal hearing, symmetric SNHL, unilateral or
asymmetric SNHL and CHL according to PTA hearing loss categories.
Excellent
(0-15 dB)
Minimal
(16-25 dB)
Mild
(26-40 dB)
Moderate
(41-55 dB)
Severe-
profound
(56-120 dB)
NH &
SHL
n
26
15
23
24
10
Age Range
(Years)
19-67
23-84
39-84
51-84
67-79
Diotic
DIN
Mean
SRT(SD)
-11.1 (0.8)
-9.7 (1.1)
-10 (1.1)
-8.7 (0.9)
-6.4 (1.5)
SE
0.16
0.28
0.22
0.19
0.49
Antiphasic
DIN
Mean
SRT(SD)
-18·4 (1.4)
-16.7 (1.6)
-15.7 (1.8)
-12.4 (2.1)
-8.2 (2.7)
SE
0.28
0.41
0.37
0.43
0.85
UHL
n
0
0
0
4
20
Age Range
(Years)
-
-
-
25-63
18-72
Diotic
DIN
Mean
SRT(SD)
-
-
-
-10.8 (0.3)
-9.4 (1.2)
SE
-
-
-
0.15
0.27
Antiphasic
DIN
Mean
SRT(SD)
-
-
-
-12.5 (1.9)
-11.3 (2.2)
SE
-
-
-
0.93
0.49
CHL
n
0
0
4
4
15
Age Range
(Years)
18-44
19-62
20-68
Diotic
DIN
Mean
SRT(SD)
-
-
-10.7 (0.7)
-9.8 (2.3)
-9.3 (1)
SE
-
-
0.35
1.12
0.27
Antiphasic
DIN
Mean
SRT(SD)
-
-
-13.7 (0.9)
-11.7 (2.4)
-10.1 (1.8)
SE
-
-
0.43
1.19
0.46
DIN: digits-in-noise, NH; Normal Hearing, SHL; Symmetric SNHL, UHL; unilateral sensorineural hearing loss, CHL;
conductive hearing loss
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
14
DISCUSSION
Antiphasic presentation improved the test characteristics of the smartphone DIN test
with higher sensitivity and specificity to detect hearing loss of various degrees, types
and symmetries than the diotic DIN. With monaural testing, it is possible to segregate
a ‘better’ ear from a ‘poorer’ ear. Traditionally, emphasis has been placed on the
function of the ‘better’ ear to assess activity and participation, but there is now
considerable evidence that asymmetric or unilateral hearing loss can reduce these
aspects of hearing health almost, or as much as symmetric binaural HL (Firszt et al.
2015; Rothpletz et al. 2012; Vannson et al. 2015). It is thus important to assess the
function of both ears, working together. Binaural tests, as used here, are more
dependent on the relative function of both ears (Supplementary Figure) but, because
of interaural summation and unmasking effects that interaction is complex (Hall et al.,
1995, 1998). A screening test should be rapid, is not intended to be diagnostic, and
persons who fail the test must be referred for diagnostic testing (Wilson and Jungner,
1968). The antiphasic DIN is a rapid test compared to sequential monaural testing and
aims to detect all hearing losses that require further diagnostic assessment.
Mechanisms of antiphasic advantage
Listeners with normal hearing in both ears were at a significant advantage for
understanding speech-in-noise compared to listeners with either type or symmetry of
hearing loss. This advantage is due to several mechanisms, but the primary one is
binaural integration. In spatial hearing, when sound from a lateral source arrives at the
nearer ear earlier than the far ear, interaural phase differences are processed as
spatial cues. Brainstem neurons detect interaural timing differences as small as 10 s
(Brughera et al. 2013), equal to about 2° of space (Middlebrooks and Green 1991). In
the antiphasic DIN, the 180° interaural phase difference of the digits simulates an
interaural timing difference, separating virtually the target speech from the noise. We
introduced a phase inversion in the speech signals between the ears, leaving the noise
in-phase (NoSπ) since the SRT improvement is larger compared to the NπSo condition
(Olsen et al. 1976). Listeners in our study with normalhearing had 6-8 dB better
antiphasic than diotic SRT, in line with the study of Smits et al. (2016). Peripheral
hearing loss disrupts interaural timing differences by desynchronizing neural activity
from the affected ear(s), reducing the antiphasic advantage (Jerger et al. 1984; Welsh
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
15
et al. 2004; Vannson et al. 2017). Predicted poorer diotic SRTs due to loss of outer
hair cell function and associated cochlear compression were also observed for
listeners with symmetric SNHL. Antiphasic SRTs, however, demonstrated greater
threshold differences in listeners with symmetric SNHL between the various categories
of hearing sensitivity compared to diotic SRT.
Unilateral hearing loss
Diotic presentation in unilateral SNHL does not result in strongly elevated SRTs
compared to listeners with bilateral normal hearing, because performance mainly
reflects the better ear. Furthermore, the 1 dB advantage provided by binaural
summation (Smits et al. 2016), was ineffective to detect unilateral SNHL. Listeners in
this study, with moderate unilateral SNHL achieved diotic SRTs comparable to
listeners with normal hearing. Similarly, diotic SRTs of those with severe-to-profound
unilateral or asymmetric SNHL compared to those with only mild symmetric SNHL.
Since listeners with unilateral SNHL could only adequately hear the digits presented
to the better ear, binaural interaction was either minimal or entirely absent. Antiphasic
SRTs were, as expected, significantly poorer and better reflected the degree of
hearing loss in the poorer ear than did diotic SRTs.
Listeners with strongly asymmetric hearing loss could increase the overall
presentation level of the DIN test by self-selecting a higher listening level. Some of
these listeners may then have enough residual hearing in the poorer ear to achieve a
degree of binaural advantage in antiphasic conditions when the signal intensity is
brought to threshold in that ear. However, the degree to which overall level adjustment
compensates for asymmetric hearing loss is also restricted to the tolerance of masking
noise in the better ear (Jerger et al. 1984). Three listeners with primarily high-
frequency unilateral SNHL had antiphasic SRTs within the normal range. Since
interaural timing differences are low frequency (< 1500 Hz) dependent (Middlebrooks
and Green 1991), it is expected that the favourable antiphasic SRTs obtained in these
three listeners was due to involvement of their residual low-frequency hearing.
Conductive hearing loss
The antiphasic test paradigm was very successful in detecting listeners with CHL. A
person with symmetric CHL could overcome loudness attenuation of the standard
diotic signals by increasing the overall presentation level, thereby achieving near-
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
16
normal standard SRTs, as seen in listeners with mild and moderate CHL. Diotic SRTs
were slightly poorer across consecutive hearing sensitivity categories (mild, moderate
and severe-to-profound), but in most cases (20/23) were still within the normal hearing
range. Earlier studies demonstrated that antiphasic processing is disrupted by acute
CHL that both attenuates and delays sound passing through the ear (Hartley and
Moore 2003). Chronic CHL commencing in infancy can impair antiphasic listening
even after CHL has resolved (Moore et al. 1991; Pillsbury et al. 1991) and produces a
number of neurological changes affecting binaural integration (Polley et al. 2013). Due
to the disruption in interaural timing difference caused by CHL, antiphasic SRTs in our
study deviated considerably from listeners with normal hearing, in contrast to diotic
SRTs.
Training and reliability
Listeners with normal hearing and SNHL had a small training effect between the
antiphasic training list and test condition. There were no significant SRT differences
between the diotic DIN and antiphasic DIN test and retest measurements. Similar
findings were reported by Smits et al. (2013), suggesting that SRT improvement from
the training list to the first test condition is due to a procedural learning effect in naïve
listeners. Overall, the antiphasic DIN test-retest reliability was high and better detected
CHL as opposed to diotic SRTs. Overall, across the entire sample in this study,
antiphasic DIN test characteristics for detecting mild and moderate hearing loss was
high. The area under the receiver operating characteristics curve for antiphasic test
accuracy for hearing losses of >25 dB HL and >40 dB HL was significantly higher (0.94
and 0.96) than for diotic testing (0.78 and 0.80).
Clinical implications
The high sensitivity of a 3-minute antiphasic DIN to detect hearing loss of various
types, symmetries and degrees holds significant potential for population-based
screening. CHL, in the form of otitis media, is typically more prevalent among
underserved, remote and poor populations than other forms of hearing loss (Hunter et
al. 2007; Cameron et al. 2014) but is not easily detected with currently used DIN tests.
Since the DIN can be used in children as young as 4 years of age (Koopmans et al.
2018), the antiphasic DIN test may be a means of early identification in those
populations, once age-specific normative SRT scores are established. School-aged
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
17
screening programs where the DIN has already been successfully implemented
(Denys et al. 2018), could similarly benefit from an antiphasic variant to improve
sensitivity and reduce test duration from a monaural to a binaural test. Of course, the
completion of a single antiphasic DIN test would not be able to differentiate between
either CHL or SNHL, or as with monaural testing, between unilateral or bilateral
hearing loss. However, following up on initial screening with other DIN variants (e.g.
monaural, filtered or modulated noise) for those who fail the antiphasic test could
potentially allow for categorization into bilateral, unilateral or CHL.
A smartphone platform of test delivery has proved a successful method of screening,
allowing for directed referrals from cloud-based data management platforms (De
Sousa et al. 2018), thereby optimizing resource allocation. Furthermore, it has been
shown that the test can be done reliably across various smartphone devices (either
iOS or Android operated) and transducers (Potgieter et al., 2016; De Sousa et al.
2018). Analysis of the hearZA tests taken approximately a year and a half after its
release, showed high test uptake (> 30 000 tests), especially among an important
target population of users younger than 40 years (De Sousa et al. 2018). The
development of the antiphasic DIN test in other language variants, however, is
recommended to make it accessible to a large global audience.
In conclusion, antiphasic SRTs correlated significantly better to poorer ear PTA than
diotic SRTs. As a result, antiphasic presentation markedly improved sensitivity to
detect SNHL and CHL, either symmetric or asymmetric, making it a powerful tool for
population-based screening.
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
18
ACKNOWLEDGEMENTS
The authors thank all the participants of this study, Steve Biko Academic Hospital and
all participating private practices for their assistance with data collection. The authors
thank Li Lin for assistance with data analysis.
This research was funded by the National Institute of Deafness and Communication
Disorders of the National Institutes of Health under award number 5R21DC016241-
02. Additional funding support was obtained from the National Research Foundation
(Grant PR_CSRP190208414782). The 2nd author’s relationship with the hearX Group
and hearZA includes equity, consulting, and potential royalties. The last author’s
relationship with the hearX Group and hearZA includes equity, consulting and potential
royalties. David Moore is supported by Cincinnati Children’s Research Foundation and
by the NIHR Manchester Biomedical Research Centre.
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
19
REFERENCES
Brughera A., Dunai, L., Hartmann, W.M. (2013). Human interaural time difference
thresholds for sine tones: the high-frequency limit. J Acoust Soc Am,133, 2839-2855.
Cameron, S., Dillon, H., Glyde, H., et al. (2014). Prevalence and remediation of spatial
processing disorder (SPD) in Indigenous children in regional Australia. Int J Audiol,
53, 326-335.
Chou, R., Dana, T., Bougatsos, C., et al. (2011). Screening adults aged 50 years or
older for hearing loss: a review of the evidence for the US preventive services task
force. Ann Int Med, 154, 347-355.
Denys, S., Hofmann, M., Luts, H., et al. (2018). School-Age Hearing Screening Based
on Speech-in-Noise Perception Using the Digit Triplet Test. Ear Hear, 39, 1104-1115.
De Sousa, K. C., Swanepoel, D.W., Moore, D.R., et al. (2018). A smartphone national
hearing test: Performance and characteristics of users. Am J Audiol, 27, 448-454.
Fellinger, J., Holzinger, D., Pollard, R. (2012). Mental health of deaf
people. Lancet, 379, 1037-1044.
Freyman, R.L., Balakrishnan, U., Helfer, K.S. (2001). Spatial release from
informational masking in speech recognition. J Acoust Soc Am, 109, 2112-2122.
Firszt, J.B., Reeder, R.M., Dwyer, N.Y., et al. (2015). Localization training results in
individuals with unilateral severe to profound hearing loss. Hear Res 319, 4855.
Genther, D. J., Frick, K. D., Chen, D., et al. (2013). Association of hearing loss with
hospitalization and burden of disease in older adults. JAMA, 309, 2322-2324.
Grant, K. W., Walden, T. C. (2013). Understanding excessive SNR loss in hearing-
impaired listeners. J Am Acad Audiol, 24, 258-273.
Hall III, J. W., Grose, J. H., Mendoza, L. L. (1995). Masker interaural phase and the
MLD: effects of conductive hearing loss. Hear Research, 84, 91-98.
Hall III, J. W., Grose, J. H., Dev, M. B., et al. (1998). The effect of masker interaural
time delay on the masking level difference in children with history of normal hearing or
history of otitis media with effusion. Ear Hear, 19, 429-433.
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
20
Hall, J. S. (2006). The development of a new English sentence in noise test and an
English number recognition test. Southampton: University of Southampton.
Hartley, D.E., Moore, D.R. (2003). Effects of conductive hearing loss on temporal
aspects of sound transmission through the ear. Hear Research,177, 53-60.
Hirsh, I. J. (1948). The influence of interaural phase on interaural summation and
inhibition. J Acoust Soc Audiol Am, 20, 536-544.
Hunter, L.L., Davey, C.S., Kohtz, A., et al. (2007). Hearing screening and middle ear
measures in American Indian infants and toddlers. Int J Pediatr Otorhinolaryngol, 71,
1429-1438.
Jansen, S., Luts, H., Wagener, K. C., et al. (2010). The French digit triplet test: A
hearing screening tool for speech intelligibility in noise. Int J Audiol, 49, 378-387.
Jansen, S., Luts, H., Dejonckere, P., et al. (2014). Exploring the sensitivity of speech-
in-noise tests for noise-induced hearing loss. Int J Audiol, 53(3), 199-205.
Jerger, J., Brown, D., Smith, S. (1984). Effect of peripheral hearing loss on the masking
level difference. Arch Otolaryngol, 110, 290-296.
Kaandorp, M. W., De Groot, A. M. B., Festen, J. M., et al. (2016). The influence of
lexical-access ability and vocabulary knowledge on measures of speech recognition
in noise. Int J Audiol, 55(3), 157-167.
Ki-moon, D.M. (2016). Hearing loss: an important global health concern. Lancet, 387,
2351-2478.
Koole, A., Nagtegaal, A. P., Homans, N. C., et al. (2016). Using the digits-in-noise test
to estimate age-related hearing loss. Ear Hear, 37, 508-513.
Koopmans, W.J., Goverts, S.T., Smits, C. (2018). Speech Recognition Abilities in
Normal-Hearing Children 4 to 12 Years of Age in Stationary and Interrupted Noise.
Ear Hear, 39, 1091-1103.
Lyzenga, J., Smits, C. (2011). Effects of Coarticulation, Prosody, and Noise Freshness
on the Intelligibility of Digit Triplets in Noise. J Am Acad Audiol, 22, 215221.
Lin, F. R., Metter, E. J., O’Brien, R. J., et al. (2011). Hearing loss and incident
dementia. Arch Neurol, 68, 214-220.
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
21
Lin, F. R., Hazzard, W. R., Blazer, D. G. (2016). Priorities for improving hearing health
care for adults: a report from the National Academies of Sciences, Engineering, and
Medicine. JAMA, 316, 819-820.
Livingston, G., Sommerlad, A., Orgeta, V., et al. (2017). Dementia prevention,
intervention, and care. Lancet, 390, 2673-2734.
Mackenzie, I., Smith A. (2009). Deafnessthe neglected and hidden disability. Ann
Trop Med Parasitol, 103, 565-571.
Middlebrooks, J.C., Green, D.M. (1991). Sound localization by human listeners. Ann
Rev Clin Psychol, 42,135-59.
Moore, D.R., Hutchings, M.E., Meyer, S.E. (1991). Binaural masking level differences
in children with a history of otitis media. Audiology, 30, 91-101.
Mulwafu, W., Ensink, R., Kuper, H., et al. (2017). Survey of ENT services in sub-
Saharan Africa: Little progress between 2009 and 2015. Glob Health Action, 10,
1289736.
Olsen, W.O., Noffsinger, D. (1976). Masking level differences for cochlear and
brainstem lesions. Ann Otol Rhinol Laryngol, 85, 820-825.
Ozimek, E., Kutzner, D., Sęk, A., et al. (2009). Development and evaluation of Polish
digit triplet test for auditory screening. Speech Comm, 51, 307-316.
Pillsbury, H.C., Grose, J.H., Hall, J.W. (1991). Otitis media with effusion in children:
binaural hearing before and after corrective surgery. Arch Otolaryngol Head Neck
Surg, 117, 718-723.
Polley, D.B., Thompson, J.H., Guo, W. (2013). Brief hearing loss disrupts binaural
integration during two early critical periods of auditory cortex development. Nat
Commun, 4, 2547.
Potgieter, J. M., Swanepoel, D. W., Myburgh, H. C., et al. (2016). Development and
validation of a smartphone-based digits-in-noise hearing test in South African
English. Int J Audiol, 55, 405-411.
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
22
Potgieter, J. M., Swanepoel, W., Myburgh, H. C., et al. (2018). The South African
English Smartphone Digits-in-Noise Hearing Test: Effect of Age, Hearing Loss, and
Speaking Competence. Ear Hear, 39, 656-663.
Reed, N. S., Altan, A., Deal, J. A., et al. (2018). Trends in health care costs and
utilization associated with untreated hearing loss over 10 years. JAMA Otolaryngol
Head Neck Surg.
Rothpletz, A.M., Wightman, F.L. and Kistler, D.J. (2012). Informational masking and
spatial hearing in listeners with and without unilateral hearing loss. J Speech Lang
Hear Res, 55, 511-531.
Ruben, R. J. (2015). How did otolaryngologyhead & neck surgery become an
essential medical discipline for the 21st century?. Int J Pediatr Otorhinolaryngol, 79,
783-785.
Smits, C., Kapteyn, T. S., Houtgast, T. (2004). Development and validation of an
automatic speech-in-noise screening test by telephone. Int J Audiol, 43, 15-28.
Smits, C., & Houtgast, T. (2005). Results from the Dutch speech-in-noise screening
test by telephone. Ear Hear, 26, 89-95.
Smits, C., Merkus, P., Houtgast, T. (2006). How we do it: The Dutch functional
hearingscreening tests by telephone and Internet. Clin Otolaryngol, 31, 436-440.
Smits, C., Theo Goverts, S., Festen, J. M. (2013). The digits-in-noise test: assessing
auditory speech recognition abilities in noise. J Acoust Soc Am, 133, 1693-1706.
Smits, C., Watson, C. S., Kidd, G. R., et al. (2016). A comparison between the Dutch
and American-English digits-in-noise (DIN) tests in normal-hearing listeners. Int J
Audiol, 55, 358365.
Smits, C. (2017). Improving the Efficiency of Speech-In-Noise Hearing Screening
Tests. Ear Hear, 38, 385-388.
Statistics South Africa. (2011). Census in brief, South Africa. Retrieved from
http://www.statssa.gov.za/publications/P03014/P030142011.pdf
The Economist. (2015). Smartphones: Planet of the Phones. Retrieved from
https://www.economist.com/leaders/2015/02/26/planet-of-the-phones
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
23
Thornton, J. L., Chevallier, K. M., Koka, K., et al. (2012). The conductive hearing loss
due to an experimentally induced middle ear effusion alters the interaural level and
time difference cues to sound location. J Assoc Res Otolaryngol, 13, 641-654.
Vannson, N., James, C.J., Fraysse, B., et al. (2017). Speech-in-noise perception in
unilateral hearing loss: Relation to pure-tone thresholds and brainstem plasticity.
Neuropsychologia,102, 135-43.
Vlaming, M. S., MacKinnon, R. C., Jansen, M., et al. (2014). Automated screening for
high-frequency hearing loss. Ear Hear, 35, 667.
Vercammen C, Goossens T, Wouters J, et al. (2018). Digit Triplet Test Hearing
Screening With Broadband and Low-Pass Filtered Noise in a Middle-Aged Population.
Ear Hear, 39, 825-828.
Vos, T., Allen, C., Arora, M., et al. (2016). Global, regional, and national incidence,
prevalence, and years lived with disability for 310 diseases and injuries, 19902015:
a systematic analysis for the Global Burden of Disease Study 2015. Lancet, 388,
1545-1602.
Watson, C. S., Kidd, G. R., Miller, J., et al. (2012). Telephone screening tests for
functionally impaired hearing: Current use in seven countries and development of a
US version. J Am Acad of Audiol, 23, 757-767.
Welsh, L.W., Rosen, L.F., Welsh, J.J., et al. (2004). Functional impairments due to
unilateral deafness. Ann Otol Rihnol, 113, 987-93.
Wilson, R. H., Civitelld, B. A., Margolish, R. H. (1985). Influence of interaural level
differences on the speech recognition masking level difference. Audiology, 24, 15-24.
Wilson, J.M.G. and Jungner, G. (1968). Principles and practice of screening for
disease. Retrieved from the World Health Organisation website:
https://apps.who.int/iris/bitstream/handle/10665/37650/WHO_PHP_34.pdf
Wilson, R. H., Zizz, C. A., Sperry, J. L. (1994). Masking-level difference for spondaic
words in 2000-msec bursts of broadband noise. J Am Acad Audiol, 5, 236-236.Wilson,
R. H., Moncrieff, D. W., Townsend, E. A., et al. (2003). Development of a 500-Hz
masking-level difference protocol for clinic use. J Am Acad Audiol, 14, 1-8.
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
24
Wilson, B. S., Tucci, D. L., Merson, M. H., et al. (2017). Global hearing health care:
new findings and perspectives. Lancet, 390, 2503-2515.
World Health Organization. (2018). Deafness and Hearing Loss Factsheet. Available
from: http://www.who.int/mediacentre/factsheets/fs300/en/
Zokoll, M. A., Wagener, K. C., Brand, T., et al. (2012). Internationally comparable
screening tests for listening in noise in several European languages: The German digit
triplet test as an optimization prototype. Int J Audiol, 51, 697-707
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
25
FIGURES
Figure 1. Antiphasic and diotic SRT according to hearing category.
Figure 2. Correlations of the diotic DIN and antiphasic DIN to poorer ear PTA.
Figure 3. ROC curves presenting test characteristics of the antiphasic DIN and diotic
for detecting poorer ear PTA > 25 dB HL (left) and > 40 dB HL (right).
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
26
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
27
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
28
All rights reserved. No reuse allowed without permission.
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/677609doi: bioRxiv preprint first posted online Jun. 21, 2019;
... Participants are instructed to attend to the target sentence while ignoring a masker. The CRM is part of a family of adaptive speech reception in noise tests (see also digit-in-noise test commonly used in audiology practice; De Sousa et al., 2019). These paradigms have been shown to be powerful tests of listening in complex environments because of their sensitivity to small intelligibility changes in highly noisy backgrounds, their applicability to testing with different maskers, and their relative independence from semantic/syntactic cues (Brungart, 2001;De Sousa et al., 2020;Eddins & Liu, 2012;Humes et al., 2017). ...
Article
Full-text available
Online recruitment platforms are increasingly used for experimental research. Crowdsourcing is associated with numerous benefits but also notable constraints, including lack of control over participants’ environment and engagement. In the context of auditory experiments, these limitations may be particularly detrimental to threshold-based tasks that require effortful listening. Here, we ask whether incorporating a performance-based monetary bonus improves speech reception performance of online participants. In two experiments, participants performed an adaptive matrix-type speech-in-noise task (where listeners select two key words out of closed sets). In Experiment 1, our results revealed worse performance in online ( N = 49) compared with in-lab ( N = 81) groups. Specifically, relative to the in-lab cohort, significantly fewer participants in the online group achieved very low thresholds. In Experiment 2 ( N = 200), we show that a monetary reward improved listeners’ thresholds to levels similar to those observed in the lab setting. Overall, the results suggest that providing a small performance-based bonus increases participants’ task engagement, facilitating a more accurate estimation of auditory ability under challenging listening conditions.
Article
Full-text available
Pure-tone threshold audiometry is currently the standard test of hearing. However, in everyday life, we are more concerned with listening to speech of moderate loudness and, specifically, listening to a particular talker against a background of other talkers. FreeHear delivers strings of three spoken digits (0–9, not 7) against a background babble via three loudspeakers placed in front and to either side of a listener. FreeHear is designed as a rapid, quantitative initial assessment of hearing using an adaptive algorithm. It is designed especially for children and for testing listeners who are using hearing devices. In this first report on FreeHear, we present developmental considerations and protocols and results of testing 100 children (4–13 years old) and 23 adults (18–30 years old). Two of the six 4 year olds and 91% of all older children completed full testing. Speech reception threshold (SRT) for digits and noise colocated at 0° or separated by 90° both improved linearly across 4 to 12 years old by 6 to 7 dB, with a further 2 dB improvement for the adults. These data suggested full maturation at approximately 15 years old SRTs at 90° digits/noise separation were better by approximately 6 dB than SRTs colocated at 0°. This spatial release from masking did not change significantly across age. Test–retest reliability was similar for children and adults (standard deviation of 2.05–2.91 dB SRT), with a mean practice improvement of 0.04–0.98 dB. FreeHear shows promise as a clinical test for both children and adults. Further trials in people with hearing impairment are ongoing.
Article
Full-text available
Purpose The smartphone digits-in-noise hearing test, called hearZA, was made available as a self-test in South Africa in March 2016. This study determined characteristics and test performance of the listeners who took the test. Method A retrospective analysis of 24,072 persons who completed a test between March 2016 and August 2017 was conducted. User characteristics, including age, English-speaking competence, and self-reported hearing difficulty, were analyzed. Regression analyses were conducted to determine predictors of the speech reception threshold. Results Overall referral rate of the hearZA test was 22.4%, and 37% of these reported a known hearing difficulty. Age distributions showed that 33.2% of listeners were ages 30 years and younger, 40.5% were between ages 31 and 50 years, and 26.4% were older than 50 years. Age, self-reported English-speaking competence, and self-reported hearing difficulty were significant predictors of the speech reception threshold. Conclusions High test uptake, particularly among younger users, and high overall referral rate indicates that the hearZA app addresses a public health need. The test also reaches target audiences, including those with self-reported hearing difficulty and those with normal hearing who should monitor their hearing ability.
Article
Full-text available
Objectives: This study aims to investigate the large-scale applicability of the Digit Triplet test (DTT) for school-age hearing screening in fifth grade elementary (5E) (9 to 12 years old) and third grade secondary (3S) (13 to 16 years old) school children. The reliability of the test is investigated as well as whether pass/fail criteria need to be corrected for training and/or age, and whether these criteria have to be refined with respect to referral rates and pure-tone audiometry results. Design: Eleven school health service centers participated in the region of Flanders (the Northern part of Belgium). Pure-tone screening tests, which are commonly used for hearing screening in school children, were replaced by the DTT. Initial pass/fail criteria were determined. Children with speech reception thresholds (SRT) of -7.2 dB signal to noise ratio (SNR) (5E) and -8.3 dB SNR (3S) or worse were referred for an audiogram and follow-up. In total, n = 3412 (5E) and n = 3617 (3S) children participated. Results: Population SRTs (±2 SD) were -9.8 (±1.8) dB SNR (5E) and -10.5 (±1.6) dB SNR (3S), and do not need correction for training and/or age. Whereas grade-specific pass/fail criteria are more appropriate, a linear regression analysis showed an improvement of 0.2 dB per year of the SRT until late adolescence. SRTs could be estimated with a within-measurement reliability of 0.6 dB. Test duration was also grade-dependent, and was 6 min 50 sec (SD = 61 sec) (5E) and 5 min 45 sec (SD = 49 sec) (3S) on average for both ears. The SRT, test reliability, and test duration were comparable across centers. With initial cut-off values, 2.9% (5E) and 3.5% (3S) of children were referred. Based on audiograms of n = 39 (5E) and n = 59 (3S) children, the diagnostic accuracy of the DTT was assessed. A peripheral hearing loss was detected in 31% (5E) and 53% (3S) of the referred children. Hearing losses found were mild. Less strict pass/fail criteria increased the diagnostic accuracy. Optimal pass/fail criteria were determined at -6.5 dB SNR (5E) and -8.1 dB SNR (3S). With these criteria, referral rates dropped to 1.3% (5E) and 2.4% (3S). Conclusions: The DTT has been implemented as the new hearing screening methodology in the Flemish school-age hearing screening program. Based on the results of this study, pass/fail criteria were determined and optimized to be used for systematic hearing screening of 5E and 3S school children. Furthermore, this study provides reference values for the DTT in children 9 to 16 years of age. Reliable SRTs can be obtained with the test, allowing accurate monitoring of hearing over time. This is important in the context of a screening guideline, which aims to identify children with noise-induced hearing loss. Validation of the screening result should go beyond taking an audiogram, as a peripheral hearing impairment cannot always be found in children with a failed test.
Article
Full-text available
Objectives: This study determined the effect of hearing loss and English-speaking competency on the South African English digits-in-noise hearing test to evaluate its suitability for use across native (N) and non-native (NN) speakers. Design: A prospective cross-sectional cohort study of N and NN English adults with and without sensorineural hearing loss compared pure-tone air conduction thresholds to the speech reception threshold (SRT) recorded with the smartphone digits-in-noise hearing test. A rating scale was used for NN English listeners' self-reported competence in speaking English. This study consisted of 454 adult listeners (164 male, 290 female; range 16 to 90 years), of whom 337 listeners had a best ear four-frequency pure-tone average (4FPTA; 0.5, 1, 2, and 4 kHz) of ≤25 dB HL. Results: A linear regression model identified three predictors of the digits-in-noise SRT, namely, 4FPTA, age, and self-reported English-speaking competence. The NN group with poor self-reported English-speaking competence (≤5/10) performed significantly (p < 0.01) poorer than the N and NN (≥6/10) groups on the digits-in-noise test. Screening characteristics of the test improved with separate cutoff values depending on English-speaking competence for the N and NN groups (≥6/10) and NN group alone (≤5/10). Logistic regression models, which include age in the analysis, showed a further improvement in sensitivity and specificity for both groups (area under the receiver operating characteristic curve, 0.962 and 0.903, respectively). Conclusions: Self-reported English-speaking competence had a significant influence on the SRT obtained with the smartphone digits-in-noise test. A logistic regression approach considering SRT, self-reported English-speaking competence, and age as predictors of best ear 4FPTA >25 dB HL showed that the test can be used as an accurate hearing screening tool for N and NN English speakers. The smartphone digits-in-noise test, therefore, allows testing in a multilingual population familiar with English digits using dynamic cutoff values that can be chosen according to self-reported English-speaking competence and age.
Article
Full-text available
Objective: Speech-in-noise hearing screening tests have become increasingly popular. These tests follow an adaptive procedure with a fixed number of presentations to estimate the speech reception threshold. The speech reception threshold is compared with an established cutoff signal to noise ratio (SNR) for a pass result or refer result. A fixed SNR procedure was developed to improve the efficiency of speech-in-noise hearing screening tests. Design: The cutoff SNR is used for all presentations in the fixed-SNR procedure. After each response a reliable test result is given (pass/refer) or an extra stimulus is presented. The efficiency and pass/refer rates between the adaptive procedure and the fixed-SNR procedure were compared. Results: An average reduction of 67% in the number of presentations can be achieved (from 25 to an average of 8.3 presentations per test). Conclusions: The fixed-SNR procedure is superior in efficiency to the adaptive procedure while having nearly equal refer and pass rates.This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.
Article
Importance Nearly 38 million individuals in the United States have untreated hearing loss, which is associated with cognitive and functional decline. National initiatives to address hearing loss are currently under way. Objective To determine whether untreated hearing loss is associated with increased health care cost and utilization on the basis of data from a claims database. Design, Setting, Participants Retrospective, propensity-matched cohort study of persons with and without untreated hearing loss based on claims for health services rendered between January 1, 1999, and December 31, 2016, from a large health insurance database. There were 154 414, 44 852, and 4728 participants at the 2-, 5-, and 10-year follow-up periods, respectively. The study was conceptualized and data were analyzed between September 2016 and November 2017. Exposures Untreated hearing loss (ie, hearing loss that has not been treated with hearing devices) was identified via claims measures. Main Outcomes and Measures Medical costs, inpatient hospitalizations, total days hospitalized, 30-day hospital readmission, emergency department visits, and days with at least 1 outpatient visit. Results Among 4728 matched adults (mean age at baseline, 61 years; 2280 women and 2448 men), untreated hearing loss was associated with 22434(9522 434 (95% CI, 18 219-$26 648) or 46% higher total health care costs over a 10-year period compared with costs for those without hearing loss. Persons with untreated hearing loss experienced more inpatient stays (incidence rate ratio, 1.47; 95% CI, 1.29-1.68) and were at greater risk for 30-day hospital readmission (relative risk, 1.44; 95% CI, 1.14-1.81) at 10 years postindex. Similar trends were observed at 2- and 5-year time points across measures. Conclusions and Relevance Older adults with untreated hearing loss experience higher health care costs and utilization patterns compared with adults without hearing loss. To further define this association, additional research on mediators, such as treatment adherence, and mitigation strategies is needed.
Article
Objectives: This study investigated whether age affects Flemish Digit Triplet Test (DTT) scores, as well as sensitivity and specificity of the test to detect high-frequency hearing impairment in middle age. Design: DTT performance in broadband and low-pass filtered noise was assessed for 21 young and 61 middle-aged persons with normal-hearing sensitivity. Cutoff values based on their performance were used to evaluate sensitivity and specificity of the DTT as a screening tool in 145 middle-aged persons, with varying audiometric profiles. Results: DTT performance for normal-hearing persons depended on age in both noise conditions. For low-pass filtered noise, the trade-off between sensitivity and specificity also increased when age-adjusted cutoff values were applied for screening in middle age. Conclusions: The Flemish DTT with broadband noise had high sensitivity and specificity to detect high-frequency hearing impairment in middle age. Similar results were obtained for low-pass filtered noise when cutoff values for referral were age adjusted.
Article
Acting now on dementia prevention, intervention, and care will vastly improve living and dying for individuals with dementia and their families, and in doing so, will transform the future for society. Dementia is the greatest global challenge for health and social care in the 21st century. It occurs mainly in people older than 65 years, so increases in numbers and costs are driven, worldwide, by increased longevity resulting from the welcome reduction in people dying prematurely. The Lancet Commission on Dementia Prevention, Intervention, and Care met to consolidate the huge strides that have been made and the emerging knowledge as to what we should do to prevent and manage dementia. Globally, about 47 million people were living with dementia in 2015, and this number is projected to triple by 2050. Dementia affects the individuals with the condition, who gradually lose their abilities, as well as their relatives and other supporters, who have to cope with seeing a family member or friend become ill and decline, while responding to their needs, such as increasing dependency and changes in behaviour. Additionally, it affects the wider society because people with dementia also require health and social care. The 2015 global cost of dementia was estimated to be US$818 billion, and this figure will continue to increase as the number of people with dementia rises. Nearly 85% of costs are related to family and social, rather than medical, care. It might be that new medical care in the future, including public health measures, could replace and possibly reduce some of this cost.
Article
In 2015, approximately half a billion people had disabling hearing loss, about 6·8% of the world's population. These numbers are substantially higher than estimates published before 2013, and point to the growing importance of hearing loss and global hearing health care. In this Review, we describe the burden of hearing loss and offer our and others' recommendations for halting and then reversing the continuing increases in this burden. Low-cost possibilities exist for prevention of hearing loss, as do unprecedented opportunities to reduce the generally high treatment costs. These possibilities and opportunities could and should be exploited. Additionally, a comprehensive worldwide initiative like VISION 2020 but for hearing could provide a focus for support and also enable and facilitate the increased efforts that are needed to reduce the burden. Success would produce major personal and societal gains, including gains that would help to fulfil the “healthy lives” and “disability inclusive” goals in the UN's new 2030 Agenda for Sustainable Development.
Article
Background: A 2009 survey of ENT, audiology, and speech therapy services and training opportunities in 18 Sub-Saharan African countries reported that the availability of services was extremely poor, the distribution of services was very inequitable, and training opportunities were limited. Objective: We conducted a new survey to determine the current status of ear, nose, and throat (ENT), audiology, and speech therapy services in sub-Saharan Africa. Method: This study is a cross-sectional study. A questionnaire was distributed by email to an ad hoc group of ENT surgeons and audiologists in 30 sub-Saharan African countries. Data from the current survey were compared to those of a 2009 survey. The numbers of ENT surgeons, audiologists, and speech therapists/100,000 people were compared to the ratios in the United Kingdom. Results: A total of 22 countries responded to the questionnaire. When data of the 15 countries that responded in both 2009 and 2015 are compared, the number of ENT surgeons had increased by 43%, audiologists had increased by 2.5%, and speech therapists by 30%. When the 23% population growth is taken into account, the numbers of ENT surgeons, audiologists, and speech therapists per 100,000 people had declined in four countries, and there remains a severe shortfall of ENT surgeons, audiologists, and speech therapists when compared to the UK Respondents cited lack of availability of basic equipment as the most frequent limitation in providing ENT services. Other important factors causing limitations in daily practice were: lack of ENT training facilities and audiological rehabilitation, low awareness of the burden of ENT pathology, as well as poor human resources management. Conclusions: There has been a lack of progress in ENT, audiology, and speech therapy services and training opportunities in sub-Saharan Africa between 2009 and 2015. There is a need to look at increased collaboration with developed countries and non-governmental organisations, establishing new and improving existing training centres in Africa, and task-shifting of some ENT services to primary health workers.