Sage Journals: Discover world-class research

Abstract

Background

Despite the proven effectiveness of traditional screening devices for Early Hearing Detection and Intervention (EHDI) programs, the costs of these instruments are high to be implemented as a sustainable solution for low- and middle-income countries. Mobile Health (mHealth) hearing screeners are an alternative solution to support EHDI programs. Therefore, demand is increasing for accessible, affordable and scalable alternatives.

Objective

This review reports on the validity, costs, status (commercially available/non-commercially available) of the mHealth screeners available globally for the paediatric population.

Method

A systematic search of PubMed, Scopus and Google Scholar was conducted. Articles published from 2014 and study populations of 0–18 years were eligible. Thirty-eight studies met the inclusion criteria, and 27 mHealth screeners were identified in this review.

Results

Three objective screeners were identified, OAEBuds, hearOAE and off-the-shelf OAE. Twenty-four subjective screeners were identified, mainly using pure-tone audiometry, game-based tests and speech-in-noise tests. Subjective screeners target children above 3 years of age. Sensitivity and specificity were comparable to those of traditional screeners, although the validation methodologies varied. Only one objective screener is commercially available. Few screeners are available free of cost and as open-source software. Compared to previous reviews on mHealth devices, this review captured objective hearing screeners.

Conclusion

The review outlined the need for more mHealth-based screeners for lower age groups, particularly for infants and toddlers, as most of the identified screeners catered to an older population of more than three years of age.

Keywords

mHealth paediatric hearing screeners good health and well-being child health innovation

Introduction

Early detection of hearing loss is critical for timely intervention through appropriate hearing device fitting and habilitation, which in turn supports optimal auditory, speech and language development.¹ Objective hearing tests, such as Otoacoustic Emissions (OAE) and Automated Auditory Brainstem Response (AABR)² are commonly used for newborn hearing screening. Hearing screening among older children at school entry or paediatric screening using pure tone audiometry and tympanometry³ are effective complementary strategies. These approaches have consistently demonstrated efficacy in ensuring the early identification and management of hearing impairment.

Despite their proven effectiveness, the widespread implementation in hospitals providing newborn hearing screening programs and in schools remains limited, particularly in low- and middle-income countries (LMICs). Access to hearing care via digital health platforms has the potential to address the rising out-of-pocket costs⁴ that have been documented for healthcare in developing countries.

Mobile Health (mHealth) is a component of eHealth and is defined as a medical or public health practice supported by mobile devices, such as mobile phones, patient monitoring devices, personal digital assistants and other wireless devices.⁵ The mHealth service delivery model offers several advantages, including accessibility, convenience, adaptability, patient-centeredness, data insights, efficiency and effectiveness.⁶ It also supports flexible administration approaches, from self-administration to screening conducted by community health workers, facilitating large-scale program implementation. The integration of health data with cloud-based data systems and electronic health records enables effective referral and follow-up processes⁷ According to a recent global report, almost 4 billion people have access to a smartphone with the internet.⁸ In India,⁹ around 94.4% of the rural population and 96% of the urban population own a smartphone, and more than 90% have access to the Internet. This widespread digital reach positions mHealth as a powerful tool for bridging healthcare delivery gaps, particularly in underserved regions. Additionally, mHealth technology eliminates the dependency on bulky hardware, offering portable handheld devices and enhanced data management systems for delivering healthcare services. Furthermore, the integration of artificial intelligence (AI) further empowers mHealth services.¹⁰

Since 2012, there has been a growing trend in the adoption of mHealth technologies for hearing care services. These services are used for health promotion, screening, diagnosis, treatment and support.¹¹ However, recent systematic and scoping reviews revealed a paucity of validated, paediatric-specific mHealth-based hearing screening tools.^11–14 Bright and Pallawella¹⁵ reviewed the validity of mHealth ear and hearing screeners available through commercial app stores. Most of the applications lacked validity in peer reviewed journals and the validity, with no evidence specific for paediatric population. Chen et al.¹⁶ performed meta-analysis to determine the accuracy of smartphone-based audiometry and speech recognition tests and reported high pooled diagnostic accuracy. However, the review primarily reflected adult populations and did not address on test paradigms for younger children. Neither review reported the availability of objective mHealth applications.

Abbey and Douglas¹⁴ and Frisby et al.¹¹ conducted comprehensive reviews of the mHealth applications available globally. Abbey and Douglas¹⁴ review did not include paediatric-specific hearing screeners and Frisby's review¹¹ mapped the use of mHealth-supported hearing care services across the continuum of care, encompassing health promotion, screening, diagnosis, treatment and support. The review also reported that hearing diagnostics and treatment applications were more available in high-income countries, whereas screening devices were more available in LMICs. Although comprehensive in scope, the reviews did not evaluate the validity of the identified mHealth applications or provide information on their commercial availability.

Hearing screening technologies can be classified based on measurement methods as objective and subjective. Objective measures assess auditory function without requiring the behavioural response of the patient/child and rely on physiological signals such as OAE or auditory evoked potentials.^17,18 Subjective measures depend on perceptual or motor responses of the patient/child to auditory stimuli. Tests such as conditional play audiometry, visual reinforcements audiometry and behavioral observational audiometry are the few examples of subjective-based hearing tests.¹⁹

In terms of service delivery, traditional hearing screening devices are stand-alone or desktop-based interfaced with a computer.²⁰ mHealth hearing screening refers to hearing screening practices supported by mobile devices such as smartphones and tablets, within this the primary interface for test administration, portability, data processing and result storage is done in the mobile device.^{7,14,15,21,22}

Therefore, this review sheds light on recent innovations and advancements in mobile phone–interfaced hearing screeners tailored for paediatric populations. Additionally, it will capture the nature of the tests (objective/subjective) and provide insights into the commercial availability and costs of these screeners. Understanding these technological developments is important, as it helps identify gaps, avoid duplication of existing efforts and guide the use of validated, cost-effective paediatric hearing screening tools into scalable solutions.

This systematic review aims to report mHealth/mobile phone or tablet-interfaced innovations that support paediatric hearing screening developed in the period of 2014–2024. Furthermore, the validity and cost of such innovations are described in detail.

Methodology

Protocol and registration

The systematic review protocol was registered in the Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD42024536576 (https://www.crd.york.ac.uk/PROSPERO/view/CRD42024536576).

This systematic review followed the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) guidelines.²³ The checklist is attached as Appendix-4 in Supplementary Material.

Search strategy

The PubMed, Scopus and Google Scholar databases were searched using a search strategy reviewed by all authors prior to the database search. These search engines were chosen based on their availability and accessibility. Additionally, manual searches of the bibliographies of the selected papers based on the eligibility criteria were performed to identify further relevant research, which was subjected to the same screening and selection process.

Search strategies were tailored for each of the above-mentioned databases, incorporating MeSH terms and Boolean operators. A pilot search was conducted in each database to refine the keywords and identify synonyms for the search strategy. The finalised search strategy was shared with all authors, and discussions on the inclusions and exclusions of the search terms were held to reach a consensus.

The details of the search strings are in supplementary file (Appendix-1). Title screening was carried out from April to June 2024; abstract screening was completed in November 2024, and the full-text search was completed in March 2025. One relevant article published during the review process was identified through hand searching and included.

Inclusion criteria

Studies using all types of designs having quantitative data were included using the PICO framework²⁴

Participants: Studies that included neonates, infants and children up to the age of 18 years.

Intervention/exposures: Studies that reported on mobile phone integrated hearing screening device/tools that can be used to test newborns, neonates, infants and children.

Comparison: Studies that report validation of the mobile phone–based tool/device in comparison with a commercial stand-alone device/tool for newborns, neonates, infants and children.

Outcomes: The primary outcomes of interest were identifying and categorizing mobile-interfaced objective and subjective hearing screening tools for the paediatric population and identifying the validity of these screening devices. Additionally, this study aimed to profile and report the commercial availability and cost of these hearing screening tools/devices.

Additionally, studies were only included if they were determined to have low bias based on the Joanna Briggs Institute (JBI) quality assessment checklist ratings (Refer Appendix-3).

Exclusion criteria: Studies which describe only diagnostic tools, Studies which are not published in English and published before January 2014.

Screening studies for eligibility

The title, abstract and full-text screening of the studies were conducted using the web-based Rayyan software (premium license was purchased). All duplicates were automatically flagged by the software, and the first author reviewed them before confirming their removal.

The titles and abstracts of the studies were screened against the inclusion and exclusion criteria by the first author (VS). After random allocation, 40% of the articles were screened by the second author (VR). Any discrepancies that arose were discussed among the reviewers, leading to a common consensus.

Two reviewers (VS and VR) screened the full text for inclusion, and any discrepancies between the decisions were discussed and resolved. Both reviewers were blinded to the screening results of the articles using Rayyan software.

Data extraction and synthesis

The data extraction categories were prepared and reviewed by all reviewers. Google Sheets were used to document full-length articles (refer to Appendix 2). The results of the extracted articles were synthesised using a narrative approach.

The cost and commercial availability of the identified screeners were manually searched on the Internet, Google Play Store and App Store. Additionally, email communications were sent to the authors to obtain further information regarding the screeners whenever necessary.

Results

Quality assessment

Following data extraction from the articles, the JBI critical appraisal checklist was utilized.²⁵ The reviewers categorised each article into three ratings: Yes, No and Unclear. If the overall appraisal of a study had two or more ‘No’ responses, then the study was considered to have ‘high bias’ and was excluded from the review. The overall bias was considered as ‘low’ if no/one item was rated as ‘unclear/high bias.’ If there were two items with ‘unclear/high bias,’ then the overall bias was considered as ‘medium,’ and if there were three or more items with ‘unclear/high bias,’ then the overall bias was considered as ‘high.’ Only studies with low and medium bias were included for data extraction; studies considered to have a high risk of bias were excluded from the review. Google Sheets was used to rate the studies. Two reviewers (VS/VR) performed a critical appraisal of the articles independently. Consensus discussions among the authors resolved any disagreements.

All studies had a cross-sectional design. Therefore, only the JBI cross-sectional quality appraisal tool was used. All the included studies for the full text were rated as having low bias and were included (refer to Appendix-3).

The search yielded a total of 4020 articles (PubMed: 3280; Scopus: 605; Google Search: 134; and hand searched: 1). During title and abstract screening, 186 duplicate articles were removed. Of the remaining 3834 articles, 3698 were eliminated for reasons such as irrelevance to the study topic, inadequate data, no validity measure and not being an mHealth technology. In total, 136 articles were selected for full-text screening.

Preferred Reporting Items for Systematic reviews and Meta-Analysis chart

Of the 136 articles included for full-text screening, 38 were considered for this systematic review. The remaining articles were excluded for reasons such as population mismatch (n = 27), lack of validation/screening focus (n = 24), inappropriate study design (n = 10), lack of mHealth technology (n = 11), inappropriate study focus (n = 20) and publication timeline beyond 2014 (n = 1). The systematic inclusion process for the articles is depicted in a PRISMA flowchart²⁶ (refer Figure 1).

Figure 1.

Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) chart.

Overview of articles included

The studies were predominantly from the USA (n = 8)^27–34 and South Africa (n = 6)^35–40. Three studies each were conducted in India,^41–43 Australia,^44–46 Brazil,^47–49 two studies each were conducted in Indonesia,^50,51 Thailand,^51,52 China,^53,54 multi country^55,56 and one each was conducted in Nicaragua,⁵⁷ Denmark,⁵⁸ Taiwan,⁵⁹ Turkey,⁶⁰ Netherlands,⁶¹ Belgium⁶² and United Kingdom.⁶³

Mobile-interfaced objective and subjective hearing screening tools for the paediatric population

Of the 38 included studies, four described objective hearing screeners, and 35 described subjective hearing screening tools/devices. However, only 27 of these tools were validated as mobile-interfaced paediatric hearing screening tools.

Classification of objective and subjective screeners

Objective hearing screeners

Number of identified screeners: Objective hearing screeners accounted for only three of the 27 validated screeners, which were only OAE-based objective tests (OAEBuds,²⁸ Off the Shelf OAE^27,29and HEAR OAE 35).

Testing site: Validation of these tools were done in hospitals (n = 2)^27,35 and quiet room (n = 2).^28,29

Test personnel: The instruments were used by the study coordinators (n = 2)^27,29 and by an audiologist (n = 1),³⁵ and one study did not mention the screening personnel involved (n = 1).²⁸

Population: The ages tested for validation ranged from birth to 32 years of age for validation. However, most of the population were newborns and infants, making it ideal for newborn hearing screenings, as they did not require any behavioural participation.

Test paradigms: Two studies described DPOAE,^27,28 one study describes TEOAE²⁹ and one study included both DPOAE and TEOAE35.

Subjective hearing screeners

Number of identified screeners: Subjective screeners accounted for most of the 24 screeners and had multiple test paradigms. These tools relied on the active participation of children (refer Table 1).

Table 1.

Categories of subjective hearing screeners.

S. no	Test category	Number of screeners & n = studies	Name of the screeners
1	Pure tone audiometry	14 screeners (n = 20)	hearScreen^31,37–40 hearcheck⁴⁷ WAHTS⁵⁷ P.E.T.I.T⁴⁸ R-App⁵⁸ Ear Scale⁵⁹ hearTest⁴⁹ hearing Test⁶⁰ Agilis Audiogram³² Hearing Threshold Test³⁴ PASS-PRO⁵² Screen H⁵¹ Shravana Mithra⁴² Smart hearing screening system⁵³
2	Game-based audiometry	4 Screeners (n = 5)	Shoebox³⁰ Sound Scouts⁴⁵ Kids Hearing Game^33,54 Hear Glue ear application⁶³
3	Speech audiometry	3 Screeners (n = 4)	Pass speech^52,64 Sound Scouts⁴⁴ SRESHT screener⁴³
4	Speech in noise	3 Screeners (n = 4)	Sound ear check⁵⁶ Kinderhoortest⁶¹ Sound scouts^44,46
5	Digits in noise	2 Screeners (n = 2)	Digit triplet test⁶² hearDigits³⁶
6	Awareness task	2 Screeners (n = 2)	Sound ear check⁵⁵ SRESHT hearing screener⁴³
7	Tone in noise	1 Screener (n = 2)	Sound Scouts^44,46
8	Behavioral observational audiometry	1 Screener (n = 1)	SRESHT Hearing Screener⁴³

Testing site: The sites of studies were done mainly in schools (n = 13),^{38,40,44–46}^{,48,52,53,56,57,59,61,62} Quiet room in community sites (n = 7)^{33,36,37,41,42,55,60} and quiet room in hospitals (n = 7).^{39,43,49,54,55,58,64} This was followed by a sound booth (n = 5),^30–32^,34,63 noise simulated at a sound booth (n = 1)⁵¹ and not mentioned (n = 1).⁴⁷

Test personnel: Most of the tests were carried by the child itself (self) (n = 18),^30,32–34^{,39,44–46,49,54–56,58,59,61–64} audiologist (n = 4),^43,48,49,53 community health worker (n = 2),^31,57 audiology student (n = 2),^37,40 ENT assistant (n = 1),⁴¹ researcher (n = 1)³⁶ and some of the studies have not mentioned the personnel (n = 6).⁴²^{47,50–52,60}

Population: The age range of the patients tested for validation was 2 days to 97 years. Because different screeners tested different age ranges, the following table shows the least age at which the screeners can screen for hearing loss (refer Table 2).

Table 2.

Least age the subjective hearing screeners can screen for hearing loss.

S. no	Age range (in years/months)	Name of the screener
1	Less than 2.11	SRESHT Hearing Screener⁴³ Hear Glue ear application⁶³
2	3–3.11	Kids Hearing Game/ Tablet hearing game⁵⁴ Smart Hearing screening system⁵³ hearScreen^31,37–40
3	4–5	Pass-pro^52,64^, hearTest⁴⁹ Kinderhoortest⁶¹ Sound Scouts^44–46 Showbox
4	6–7	Hearing Threshold Test³⁴ KHG Tablet application³³ P.E.T.I.T⁴⁸ WAHTS⁵⁷ Ouviu & HearCheck⁴⁷ hearDigits³⁶
5	8- 10	Screen H (screen out)^50,51 Digit triplet test⁶² Shravana Mithra⁴²
6	11 and above	R- App (Research app)⁵⁸ Ear scale⁵⁹ Agilis Audiogram³²

These tools are predominantly aimed at preschool and school-aged children, enabling mass screening in community, school or home environments. Unlike objective screeners, subjective screeners require the active participation of children, and most screeners can only validate or use screeners for children above three years of age.

Commercial availability of hearing screeners

The commercial availability of the identified 27 mHealth screeners was assessed. For the objective hearing screener hearOAE is known to be available commercially, two studies^27,29 which described off the shelf OAE, is available as an open source and therefore reported only the manufacturing cost. The OAEBuds²⁸ are not commercially available but have reported the costs for manufacturing, hence both off the shelf OAE and OAEBuds were considered as prototypes.

Among the subjective hearing screeners (n = 24), 15 screeners were commercially available, 1 (Shoebox) had discontinued services for the paediatric population, and no information on eight screeners’ commercial availability.

The cost profiles of the screeners were compiled based on their availability in public search engines. Efforts were also made to contact the authors/vendors to obtain a quotation for the screener (refer Table 3).

Table 3.

Cost and commercial availability of the screeners.

Name of the screener	Cost in USD	Source of data on commercial availability
Objective
Hear OAE	$2300 (Yearly License- $350)	https://hearxgroup.com/products/hearoae
OAEBuds	$28.30	Not available (Prototype)
Off the shelf OAE	$10 (manufacturing cost)	Not applicable (Prototype)
Subjective
hearScreen	Lite- $1200 (database- $370/ year). Enterprise- $1500 (Database-$680/year), Enterprise plus- 1500 (Database-$825/year)	https://hearxgroup.com/products/hearscreen
Shoebox	discontinued services for paediatrics	https://www.shoebox.md/
hearDigits	$995/ year	https://hearxgroup.com/products/heardigits
Pass- Pro	Free of cost	http://play.google.com/store/apps/details?id = com.facgure.screening&hl = hi
WAHTS	$7350 + 1750 (software license and calibration fee)	https://wahtshearing.com/#wahts-system
Ear- scale	Free of cost	https://apps.apple.com/in/app/ear-scale/id1401194418
Kids Hearing Game/ Tablet Hearing game	Free of cost	https://apps.apple.com/us/app/kids-hearing-game/id1193370040
Hearing test	Free of cost	https://play.google.com/store/apps/details?id = mobile.eaudiologia&hl = en_IN&pli=1
Kinderhoortest	Free of cost	https://www.kinderhoortest.nl/
Hearing threshold test	The stand-alone price is not accessible. Total NIH toolbox which includes cognition, Motor, Emotion and Sensation tests are available.1–2 Devices: $599.99 3–6 Devices: $1499.99 7–10 Devices: $2499.99	https://nihtoolbox.org/get-the-toolbox/
Digit triplet test	Set up cost- 1000 euros + license fee (N/A)	https://www.astp4kt.eu/about-us/kt-news/digit-triplet-test-screening-for-hearing-loss-made-swift-and-easy.html
Hear glue ear application	Free of cost	https://www.hearglueear.co.uk/hear-glue-ear
Screen H (screen out)	Not available	www.Screenout.id
Shravana Mithra	Software free, Transducer cost	https://apkcombo.com/jss-shravana-mitra/com.hearingaid.jsstu.HearingAid/
SRESHT hearing screener (commercially known as ASPHEAR-Hear Kids)	$ 349.86 + $ 69.97/year	https://play.google.com/store/apps/details?id = com.chakra.sreshthearkids
hearTest	lite- $2200 + $430/year Enterprise- $2500 + $760/ year Enterprise plus- $2500 + $900/year	https://hearxgroup.com/products/heartest
Sound scouts	4 tests – $34.99	https://www.soundscouts.com/download/
Ouviu & Hearcheck	Not available	Not available
P.E.T.I.T	Not available	Not available
R-App (research app)	Not available	Not available
Sound ear check	Not available	Not available
Smart hearing screening system	Not available	Not available
KHG tablet application	Not available	Not available
Agilis audiogram	Not available	Not available

Validity of the screening tools

Different validity measures were used for the mHealth screeners. For objective hearing screeners, sensitivity and specificity analyses (n = 3) were validated against commercial OAE screeners^27–29 and concordance (n = 1), validated against the diagnostic OAE instrument³⁵ were done). For subjective hearing screeners, sensitivity and specificity (n = 21), Cohen's kappa (n = 1), Pearson's (n = 3) and Spearman rank correlation (n = 1), absolute value comparisons (n = 2) and mean values (n = 3) (refer to Tables 4 and 5).

Table 4.

Validity of objective hearing screeners.

S. no	Name of the screener	Type of test	Number of validation studies	Age range	Sample size for children	Sample size for adults	Pass /refer criteria	Validity	Reference standard
1	OAEBuds	TEOAE	1²⁸	7–32 years	12	14	TEOAE – of Any 2-frequency exceeding 8 dB SNR & Emission amplitude greater than −10 dB SPL	TEOAE – sensitivity of 100.0% (95% CI, 64.6–100.0%) and speciﬁcity of 89.7% (95% CI, 80.2–94.9%). Using two frequency bands as the pass criteria yielded an Area Under the Curve (AUC) of 0.958. Using three or four frequency bands as the pass criteria results in AUCs of 0.950 and 0.884 respectively	DPOAE and TEOAE screener
2	Hear OAE	DPOAE and TEOAE	1³⁵	mean age – 4.5 days (SD: 11.3)	176	-	DPOAE – is an SNR of greater than or equal to 6 dB for four out of six test frequencies in both screening devices. TEOAE – SNR of 3 dB or more in three out of five frequencies	DPOAE is concordance level – 89.7% where noise levels are significantly lower than compared OAE diagnostic instrument TEOAE concordance 85% which had significantly larger SNRs	Otodynamics ILO OAE (DP & TE OAE)
3	Off the shelf OAE	DPOAE	2^27,29	Week-20 years	201 ears and 74 infants	-	DPOAE – 5 dB SNR or more in any of the three or more frequencies	Ali et al.²⁷ – Sensitivity – 100% (95% CI, 0.69–1.00), Specificity – 88% (95% CI, 0.78–0.95), PPV – 56% (95% CI, 0.31–0.78), NPV – 100% (95% CI, 0.94–1.00). Chan et al.²⁹ – Sensitivity of 100% (95% CI, 94.5–100.0%) and Specificity – 88.1% (95% CI, 81.6–93.5%).	Commercial OAE screeners

Table 5.

Validity of subjective hearing screeners.

S. no	Name of the screener	Type of test	Number of validation studies	Age range	Sample size for children	Sample size for adults	Pass /refer criteria	Validity	Reference standard
1	hearScreen	Pure tone audiometry	5^31,37–40	3–97	1489	1110	Pass- < 25 dBHL in each of the four test frequencies	Swanepoel et al.⁴⁰ The sensitivity and specificity were calculated manually using cross tabs which yielded a Sensitivity of 62.5% and specificity of 98.7%. Phanguphangu et al.³⁹ reported a Sensitivity of 82% and specificity of 94, positive predictive value 90% and Negative Predictive Value of 88% for children with HIV/AIDS. Additionally, the test–retest value was analysed, and Pearson's correlation coefficient values of r = 1 for normal hearing and r = 0.92 for hearing impairment were obtained. Manayan et al.³¹ reported a sensitivity of 85% and specificity of 41%, Positive Predictive Value (PPV) of 36% and Negative Predictive Value (NPV) – 87%. Mahomed-Asmail et al.³⁸ reported a sensitivity of 75%, and specificity of 98.5%, PPV of 52.9% and NPV of 99.4%. Louw et al.³⁷ reported the Sensitivity as 81.7% and Specificity of 83.1%.	Screening – Screening audiometers Diagnostics – Standard pure tone audiometry (AC and BC)
2	Shoebox	Automated game like play audiometry	2^30,41	5–18	102	–	Pass-≤15 dBHL	Pereira et al.³⁰ Sensitivity was not reported because of less sample size. Specificity for ≥16 dB is 37% * for ≥21 dB HL was 52% Yalamanchali et al.⁴¹ – Sensitivity – 89%, Specificity – 70%	Diagnostics Pure tone audiometry/Air conduction audiometry for 05, 1, 2, 4, 6 and 8 kHz
3	hearDigits	Digits in noise	1³⁶	6–13	612	–	50% of digits recognized correctly	Sensitivity and specificity is not calculated, only age wise normative scores have been done	Inclusion criteria is normal hearing children which was done using otoscopy, tympanometry and a screening audiometer
4	Ouviu & HearCheck	Automated pure tone audiometry	1⁴⁷	6–96	38	147	Pass-1 K, 2 K and 4K- < 20 dBHL and 500 Hz and 8 khz- < 40 Hz.	Ouviu – Sensitivity (97.1%), Specificity (96.6%), PPV (94.3%). NPV (98.3%) Hearcheck sensitivity (79%), specificity (98.3%), PPV (96.4). NPV (89.1)	Retrospective pure tone audiometry data
5	WAHTS	Tablet-based audiometry screening	1⁵⁷	7–9 years	3398	–	Pass- < 25 dB HL	Specificity – 99.6%, PPV – 77%	Otoscopy, tympanometry, pure tone audiometry
6	P.E.T.I.T	Automated pure tone audiometry	1⁴⁸	6–12	244	–	Pass- 1 K, 2 K and 4K- < 20 dBHL and 500 Hz– < 30 Hz.	Sensitivity (95.7%), specificity (81%), PPV (34.4%). NPV (99.4%).	Pure tone audiometer and tympanometry
7	R-App (Research app)	Automated pure tone audiometry	1⁵⁸	11–85	3	55	Pass- < 25 dB HL	Compared with manual audiometry – Absolute difference across frequency is between 4.5 and 6.2 dB	Pure tone audiometry
8	Sound ear check	Speech in noise	2^55,56	5–9.4	870	–	Identifies 71% correctly (Noise fixed at 65 dB SPL but stimuli are up down method – 2 dB steps)	Sensitivity was calculated for different types of HL (CHL and SNHL). SEC (ref) Sensitivity is 63%, 72% and 89% for 20-, 30- and 40-dB HL for combined types of HL, respectively, Specificity is 58, 68 and 79 for 20-, 30- and 40-dB reference tests, respectively. For SEC (aph) the sensitivity is 63%, 69% and 80% for 20, 30 and 40 dB for combined types of different hearing losses, respectively, specificity for the same is 67%, 65% and 76%, respectively	Pure tone audiometry for 500 Hz to 4 kHz
9	Pass-pro	Speech audiometry and pure tone audiometry	2^52,64	4–5	107	–	2/3 correct response for each presenting intensities for each ear (40, 30 and 20 dB)	Sensitivity and Specificity for TDH39 headphones. Normal hearing: 37.5% and 81.9%; mild hearing loss: 50% and 88.1%; Moderate Hearing loss: 100% and 92%. Beyerdynamic DT 770 PRO headphones Normal hearing – 25%, 81.9%, Mild hearing loss – 50%, 92.9%, Moderate Hearing loss – 100%, 94.3%. Generic headphones Normal hearing – 50%, 75%, Mild hearing loss – 50%, 91.7%, Moderate Hearing loss – 100%, 93.1%	Standard conditioned play audiometry for 500 Hz to 4 KHz
10	Ear scale	Pure tone audiometry	1⁵⁹	11–12	85	–	Pass- < 25 dB HL in both ears	Sensitivity – 100%, Specificity – 95.2%	Pure tone audiometry for 500 Hz to 4 kHz
11	Kids Hearing Game/ Tablet hearing game	Play-based pure tone audiometry	1⁵⁴	3–13	516	–	Pass- < 20 dBHL in all frequencies	Above 4 years: Sensitivity – 91% and Specificity – 74%, below 4 years: Sensitivity – 100% and Specificity – 29%	Pure tone air conduction audiometry from 250 to 8 kHz
12	Hearing test	Pure tone audiometry	1⁶⁰	5–15	50	–	Not mentioned but classified the hearing loss under normal, mild and moderate	kappa coefficient value of the 100 ears was 0.059. The kappa coefficient values for the conventional and smartphone audiometry data at each frequency (500, 1000, 2000 and 4000 Hz) were 0.036, 0.058, 0.052 and 0.010, respectively. All p-values were >0.05	Pure tone audiometry, otoscopy and tympanometry
13	Smart hearing screening system	Pure tone audiometry	1⁵³	3–6	6288	–	Refer- > 25 dBHL for average of 500 Hz, 1 K, 2 K and 4 kHz	5% of the population (312 children) were tested using PTA, Tympanometry and DPOAE prior to screening. 37.5% sensitivity and 92.6% specificity	Pure tone/ play-based audiometry with tympanometry and DPOAE
14	Kinderhoortest	Speech in noise with Picture Identification task	1⁶¹	5–12	94	–	Speech reception threshold (SRT in dB SNR) at which 50% of the material is correctly understood, was based on the mean SNR values of the last 10 presentations	Only age-related mean scores of the SRT were identified, and not sensitivity and specificity. Regression analysis reported a significant improvement of 0.3 dB SNR in mean SRT score per age year (β = 0.25, 95% CI: −0.43 to −0.08; p = 0.004, R² = 0.11)	Pure tone audiometry at 500, 1, 2 and 4 kHz
15	KHG tablet application	Game-based audiometry	1³³	6–11	11	–	Pass- < 25 dBHL	Mean results were compared using the test of equivalence. Statistically equivalent values were obtained for the tablet and the conventional audiometry (p < 0.050) except for 500 Hz (p = 0.101)	Pure tone audiometry
16	Hearing threshold test	Pure tone audiometry	1³⁴	6–76	27	63	Threshold tracking for 500, 1 K, 2 K and 4 kHz (WHO-based classification HL if better ear PTA > 20 dBHL	The overall sensitivity was 96% and the specificity was 96%	Pure tone audiometry
17	Digit triplet test	Speech in noise	1⁶²	9–16	7029	–	The pass/fail criteria were determined at −7.2 dB SNR in 5E and −8.5 dB SNR in 3S	Pearson correlation Strong correlation- r = 0.65 (left ear) and moderate correlation r = 0.59 (right ear)	Pure tone audiometry
18	Hear glue ear application	Game-based audiometry	1⁶³	2–8	60	–	Standard diagnostic pure tone audiometry classifications	Spearman rank correlation r = −0.656- Strong correlation	Pure tone audiometry
19	Screen H (screen out)	Pure tone audiometry	2^50,51	8–84	37	417	Pass- < 25 dBHL	Rahim et al.⁵¹ – Strong correlation = r = 0.843–0.899 and Sen – 90.9%, Spec – 98.3%, accuracy – 93.6%, PPV – 99.4%, NPV – 84.8% Sudiro et al.⁵⁰ – sensitivity (80.8%–82.8%) and specificity (93.9%–94.1%)	Pure tone audiometry
20	Shravana Mithra	Pure tone audiometry	1⁴²	10–60	87	693	Standard pure tone audiometry classification	AC values – 99% within 10 dB and BC – 97% of the values obtained from PTA. No sensitivity and specificity analysis done	Pure tone audiometry
21	SRESHT Hearing Screener	Behavioral Observational Audiometry (BOA), Speech Spectrum Awareness Test (SSAT), Speech Recognition Test (SRT)	1⁴³	2 days–6 years	55	–	BOA and SSAT-Response at 60 dB HL-Pass SRT: Correct response of 5/6 words in each ear (80% psychometric function)	kappa overall value: 82% (CI: 0.66–0.97)	Pure tone audiometry
22	HearTest	Pure tone audiometry	1⁴⁹	5–92	1110	300	Children Pass- < 30 dBHL, adults <40 dBHL	Corona et al.⁴⁹ – Children (n = 26 children/56 ears) – Disabling HL-Self-operated Sensitivity – 100% Specificity – 98%, tester operated sensitivity – 100% specificity – 94.1%. Any type of HL-Self operated: Sensitivity – 100% Specificity – 90.2%, Tester operated: Sensitivity – 100%, Specificity – 84.3%	Pure tone audiometry- Screening and diagnostics
23	Agilis Audiogram	Pure tone audiometry	1³²	12–28	7	20	Not mentioned	For the paediatric population, the mean difference was in the range of −8.8 (8 kHz) to 6.5 (4 kHz) (Pearson correlation = 0.93; p < 0.001	Pure tone audiometry
24	Sound scouts	Game-based pure tone audiometry, speech-in-quiet test, a speech-in-noise test and a tone-in-noise test	3^44–46	4–15	1766	20	Pass-H metric of −1.8 SRT, Refer −2.5 SRT and four frequency average – (500, 1 K, 2 K and 4 kHz) > 30 dBHL – Hearing Loss	Dillon et al.⁴⁶ – Only the overall validity of the tool was reported with a reported sensitivity of 85% and specificity of 98% when inconclusive results were excluded Bowers et al.⁴⁴ – Sensitivity not reported, Specificity – 94.99% (false positive on rescreen was 5.01%) Mealings et al.⁴⁵ – The sensitivity of Sound Scouts for a 4-FAHL of >20 dB HL was 41%, and the specificity was 89%. Sensitivity was improved when using hearing loss criteria 4-FAHL of >30 dB HL, with a sensitivity of 88% and specificity of 88%.	Otoscopy, immittance and pure tone audiometry

The validity of all screeners was categorised based on their commercial availability.

Commercially available validated mHealth Screeners

Objective hearing screeners:

Only hearOAE was available commercially, and it contained two test categories: DPOAE, which follows a pass criterion of an SNR greater than 6 dB in four out of six frequencies. For TEOAE a SNR of 3 dB or more in three out of five frequencies.³⁵ The study reported a concordance level of 89.7% and TEOAE of 85%.³⁵

Subjective hearing screeners

Since subjective hearing screeners have a variety of test paradigms, the validity is described in separate headings. The pass criteria were subjective to the screeners. Hence, based on this criterion, validity is reported.

Pure tone audiometry (n = 8)

hearScreen, WAHTS, PASS-PRO, Ear Scale, Hearing Test, Hearing Threshold Test, Shravana Mithra and hearTest are commercially available screeners. The screeners were validated against diagnostic pure-tone audiometry. Some screeners compared the results with otoscopy and tympanometry.

hearTest⁴⁹ – The pass criterion is <30 dB HL- Sensitivity is 100% and specificity is 98% for disabling hearing loss, and sensitivity is 100% and specificity is 90.2% for any type of hearing loss.

WAHTS,⁵⁷ hearScreen,^37–40 Ear Scale⁵⁹ and Pass-pro⁵² – The pass criterion is <25 dBHL-WAHTS⁵⁷ reported only the specificity (99.6%) and positive predictive value (77%). For the other screeners, the sensitivity ranged from 62.5%⁴⁰ to 100%⁵⁹ and a specificity range of 81.9%⁵² to 98.5%.³⁸

hearScreen³¹ and Hearing Threshold test³⁴ – The pass criterion was <20 dB HL. hearScreen³¹ reported a sensitivity of 85% and specificity of 41%. Hearing Threshold test³⁴ reported to have a sensitivity and specificity of 96%.

Hearing Test⁶⁰ and Shravana Mithra⁴² – The pass criterion was <15 dB HL. Hearing Test⁶⁰ reported a kappa coefficient value of 0.059 and Shravana Mithra⁴² reported the absolute values for air conduction testing is 99% and bone conduction testing is 97% and these are within the 10 dB when compared to standard pure tone audiometry testing.

Game-based audiometry (n = 3)

Kids hearing game,^33,54 Hear Glue ear application⁶³ and Sound scouts⁴⁵ are commercially available game-based audiometry:

Kids Hearing Game/ Tablet hearing game – The pass criterion is <20 dBHL in all frequencies. Above 4 years – Sensitivity – 91% and Specificity – 74%, below 4 years – sensitivity is 100% and specificity is 29%⁵⁴ and mean results were statistically equivalent (p < 0.050) except for 500 Hz (p = 0.101).³³

Sound scouts: Pass criterion – four frequency average – (500, 1 K, 2 K and 4 kHz) < 30 dB HL. The sensitivity and specificity were reported to be 88%.⁴⁵

Hear Glue ear application: The pass/refer criterion was not mentioned however the stimuli of presentation was in the range of 20–80 dB HL and was compared with previous diagnostic thresholds. The validity is measured through Spearman rank correlation r = −0.656, which indicates a strong correlation.

Speech audiometry (n = 3)

The SRESHT Hearing screener,⁴³ Pass-pro (n = 2)^52,64 and Sound scouts.⁴⁶ All screeners compared their results with the gold standard pure tone audiometry tests:

Pass-speech^52,64 used a 50% psychometric function and the presentation of stimuli was 20 dB HL, with a reported sensitivity is 90% and the specificity is 93.4%.

Sound scouts⁴⁶ used a H metric of −1.8 SRT, Refer- (-)2.5 SRT with reference to pure tone values. Only the overall validity of the tool was reported with 85% sensitivity and 98% specificity after inconclusive results were excluded.

SRESHT Hearing screener⁴³ used the correct response of 5/6 words in each ear (80% psychometric function) as the pass criterion which reportedly yielded a kappa overall value of the screener of 82%.

Speech in noise (n = 2)

Kinderhoortest⁶¹ and Sound scouts⁴⁴ are the commercially available screeners for speech in noise. Kinderhoortest⁶¹ and Sound scouts⁴⁴ compared the screeners using gold standard pure tone audiometry. Sound scouts⁴⁴ additionally used otoscopy and tympanometry results.

Sound scouts: The scoring criteria are as follows: Fail (0–70), borderline (71–78.4), children with a four-frequency average hearing loss (4 FAHL; average of 0.5, 1, 2 and 4 kHz) of 30 dB HL in at least one ear will reliably fail or borderline score. Only the overall validity of the tool was reported with 85% sensitivity and 98% specificity after inconclusive results were excluded⁴⁶ and in another article the reported specificity is 94.99% (false positive on rescreen was 5.01%)⁴⁴

Kinderhoortest: The pass criterion speech reception threshold (SRT in dB SNR) at which 50% of the material was correctly understood. The age-related mean value of the scores was reported as – 14.4 dB SNR (SD: 1.6). They also reported that the presentation type (computer or smartphone) did not affect test scores.

Digits in noise (n = 2)

Digit triplet test⁶² and hearDigits³⁶ are commercially available. The screeners’ validity measures were compared with pure tone audiometry, and hearDigits additionally used otoscopy and tympanometry:

Digit triplet test⁶² – The pass criterion is better than −7.2 dB SNR for younger children (9–12 years old) and −8.5 dB SNR for older children (13–16 years old). The validity measure used was Pearson correlation which revealed a strong correlation of r = 0.65 (left ear) and moderate correlation r = 0.59 (right ear).

hearDigits³⁶ – The pass criterion is 50% of digits recognized correctly and the age wise normality of the screener was calculated diotic SRT mean value of dB SNR is −9.2 (minimum −12.4 and maximum −6.8 dB SNR) and antiphasic value of dB SNR is −16.4 dB SNR (minimum −20.6 dB SNR and maximum −11.2 dB SNR) SRT mean the SRTs improved by 0.15 dB and 0.35 dB SNR per year of age.

Awareness tasks (n = 1)

The SRESHT Hearing Screener is the only commercially available screener. It uses the Speech Spectrum Awareness Task to screen for disabling hearing loss (>60 dBHL). SRESHT Hearing screener⁴³ was validated against pure tone audiometry.

1. SRESHT Hearing Screener Criterion: Three out of four stimuli are considered a pass for 60 dBHL. Only the overall kappa value for the screener was available (82%).

Tone in noise (n = 1)

Only sound scouts had two studies discussing tone in noise^44,46 and commercially available. They had multiple tests, such as a speech-in-quiet test, speech-in-noise test and tone-in-noise test, combined to give an overall score. These results were compared with the gold standard pure tone audiometry⁴⁴ and were additionally tested with an otoscope and tympanometry.

Both studies used different metrics to refer to children. Dillon et al.⁴⁶ used the H metric (which is used as a reference of normal limits, a lower H metric indicated better speech perception in lower SNRs, and higher H metric indicated poor performance at lower dB SNRs). A score was considered a Pass if the H metric was −1.8 SRT and classified as a refer if it was −2.5 SRT. Bowers et al.⁴⁴ had a cumulative scoring pattern of the screener and classified the metrics as follows: It is considered as fail (0–70), a borderline (71–78.4) and a pass (79–145+). Children with a 4-FAHL (average of 0.5, 1, 2 and 4 kHz) of 30 dBHL in at least one ear will reliably return as fail or borderline score. Only the overall validity of the tool was reported with a reported sensitivity of 85% and specificity of 98% when inconclusive results were excluded, as reported by Dillon et al.⁴⁶ and Bowers et al.⁴⁴ reported the overall specificity of the Sound Scouts screener as 94.99%, which was based on a 5.01% false positive rate.

Behavioural observation audiometry (n = 1)

The SRESHT hearing screener⁴³ explored behavioural observation audiometry as a screening module for infants and toddlers aged 0–1 year with a mean age of 6 months for children with normal hearing and a mean hearing age of 8 months for children with hearing impairment. The screener was compared with pure tone audiometry, and subjective responses such as eyeblink, eye widening, searching and head turning towards the sound source were considered as a response. If the infant/toddler can respond to ¾ stimuli correctly, it is considered a pass. Module-specific validity was not available; the overall kappa value was 82%.

Among the commercially available tools (refer Table 6), the hearTest, Ear Scale and Hearing Threshold Test demonstrated the highest sensitivity for pure-tone-based screening, whereas Sound Scouts consistently showed acceptable sensitivity across multiple subjective paradigms, including game-based and speech-in-noise screening.

2. Non-commercially available/prototype validated mHealth screeners

Table 6.

Summary of commercially available screeners based on higher validity.

S. no	Screening paradigm	Screener	Age	Sensitivity reported
Objective
1.	DPOAE & TEOAE	hearOAE	Newborns and infants	High concordance (DPOAE 89.7%, TEOAE 85%)
Subjective
1.	Pure tone audiometry	hearTest	4 years	Sensitivity is 100% and specificity is 98%
		Ear Scale	11 years	Sensitivity is 100% and Specificity is 95.2%
		Hearing Threshold Test	6 years	Sensitivity and specificity is 96%
2.	Game-based audiometry	Sound Scouts	4 years	Sensitivity and specificity is 88%
3.	Speech audiometry	Pass-Speech	4 years	The sensitivity and specificity were 90% and 93.4%, respectively.
4.	Speech in noise	Sound Scouts	4 years	Sensitivity is 85% and specificity is 98%
5.	Awareness task	SRESHT Hearing Screener	1–3 years	Overall kappa value 82%
6.	Behavioral observational audiometry	SRESHT Hearing Screener	0–1 year	Overall kappa value 82%

Objective hearing screeners

Both OAEBuds²⁸ and Off the Shelf OAE^27,29 were published as open-source codes and hence, they are not commercially available and are validated as prototypes.

The OAEBuds²⁸ contain TEOAE screening and Off the shelf OAE^27,29 contains DPOAE screening both these devices were validated against commercial OAE screeners.

Off the Shelf OAE^27,29: The pass criteria for DPOAEs required an SNR greater than 5 in three out of four frequencies.^27,29 The reported sensitivity is 100%^27,29 and specificity of 88%

OAEBuds²⁸: For TEOAEs, the criteria included any two frequencies exceeding 8 dB SNR and an emission amplitude greater than −10 dB SPL. The studies reported a sensitivity of 100% and specificity of 89.7%.²⁸

Subjective hearing screeners:

Pure tone audiometry (n = 7): All the screeners’ results were compared with the gold standard pure tone audiometry, and some of the screeners were additionally compared with otoscopy, tympanometry and ENT diagnosis. Shoebox has discontinued its services for paediatric populations; however, since the tool is validated, it is being reported here.

Screen-H⁵⁰: The pass criterion was <35 dB HL,⁵⁰ the range of sensitivity (80.8%–82.8%) and specificity (93.9%–94.1%).

R-App⁵⁸: The pass criterion was <25 dBHL. The absolute difference of dB between the screener and a manual audiometer is between 4.5 and 6.2 dBHL.

Smart hearing screening⁵³ and Screen-H⁵⁰: The pass criterion was < 25 dB HL, the range of sensitivity (37.5%–82.8%) and specificity (92.6%%–98.5%).

Ouviu,⁴⁷ P.E.T.I.T⁴⁸: The pass criterion was <20 dB HL. For 500 Hz, the pass criteria were <30 dB HL⁴⁸ and <40 dB HL.⁴⁷ The sensitivity is in the range of 95.7%⁴⁸ −97.1%⁴⁷ and specificity is in the range of 81%⁴⁸–96.6%.⁴⁷

Shoebox⁴¹ and Agilis Audiogram³²: The pass criterion was <15 dB HL. For Shoebox⁴¹ the reported sensitivity was 89% and specificity was 70%. For the Agilis Audiogram,³² a Pearson correlation value of 0.9 has been reported.

Game-based audiometry (n = 1): Shoebox is validated against Air conduction audiometry for the frequencies of 0.5, 1, 2, 4, 6 and 8 kHz:

Shoebox: Sensitivity was not reported because of the small sample size. The specificity for ≥16 dB was 37% and for ≥ 21 dB HL was 52%.³⁰

Speech in noise (n = 1): The Sound ear check⁵⁶ was validated against diagnostic pure tone audiometry.

1. Sound ear check criterion: Pass-Standard deviation of more than 2.5 dB SNR for the last 15 trials. The overall sensitivity and specificity of the screener for different degrees of hearing loss were <20–69% and 65%, <30–79% and 78%, <40 dB – 90% and 81%, respectively.

Awareness tasks (n = 1): The Sound ear check⁵⁵ used pure tone audiometry tests as a comparison to the screener.

Sound ear check criterion: Phase I- Sound acclimatisation's- Dichotic presentation, Noise fixed at 65 dB SPL @ 0 dBSNR, done until correctly identified. Phase-II Training (21 trials) Identifies 71% correctly (Noise fixed at 65 dB SPL but stimuli is up down method – 2 dB steps). SEC (ref) test Sensitivity was 63%, 72% and 89% for 20, 30 and 40 dB HL, respectively, and specificity was 58%, 68%, 79% for 20-, 30- and 40-dB HL, respectively. For the SEC (aph) test, the sensitivity was 63%, 69% and 80% for 20, 30 and 40 dB, and the specificity was 67%, 65% and 76%, respectively, for the combined types of different hearing losses.

Prototype and non-commercial objective hearing screening tools (refer Table 7), such as OAEBuds and Off the Shelf OAE, demonstrated good sensitivity, comparable to or exceeding that of commercially available devices. However, their lack of market availability limits immediate scalability, reinforcing the need to distinguish between technical validity and implementation readiness.

Table 7.

Summary of non-commercially available/prototype screeners based on higher validity.

S. no	Screening paradigm	Screener	Age	Sensitivity reported
Objective
1.	DPOAE	Off the shelf OAE	Newborns and infants	Sensitivity is 100% and specificity is 88%
2.	TEOAE	OAEBuds	7–32 years	Sensitivity of 100% and specificity of 89.7%
Subjective
1.	Pure tone audiometry	Ouviu	6 years	Sensitivity is 97.1% and specificity is 96.6%
1.	Pure tone audiometry	Screen-H	8 years	Sensitivity is 90.9% and specificity is 98.3%
2.	Game-based audiometry	Shoe box	6 years	The sensitivity was not reported. Specificity for ≥16 dB was 37% and for ≥21 dB HL was 52%

Similarly, subjective mHealth screeners demonstrated high to acceptable sensitivity and strong agreement with diagnostic pure-tone audiometry across multiple test paradigms. Nonetheless, heterogeneity in validation metrics, pass criteria and the absence of commercial availability limits their immediate applicability beyond research settings.

The compliance of all validity data is available in Table 4 for objective mHealth screeners and Table 5 for subjective mHealth screeners.

Hardware and transducers used

The studies included using mHealth screeners interfaced with different hardware to present stimuli. In Objective screeners, OAEBuds²⁸ innovatively used a wireless technology, while others used wired probe connected to a controller interfaced with a smartphone for display.

For Subjective hearing screens, most of the studies use tablets and smartphones interfaced with standard headphones. There were a few articles that used off the shelf headphones for presentation of stimuli^{42,59–6133,43,51,53,54} (n = 9) (refer Table 8).

Table 8.

Classification of hardware interfaced transducers.

Objective:	Number of studies
Wireless interfaced with smartphone	1²⁸
Probe interfaced with smartphone	3^27,29,35
Subjective	Number of studies
Smartphone interfaced with standard transducers	11 ^31,36–40 ^{,45,49,50,52,58}
Software using phones transducers	1 ⁴⁷
Tablet mobile phone interfaced with standard transducers	14 ^{30,32,34,41,44–46} ^{,48,50,55–57,62–64}
Smartphone with off-shelf transducers	4^42,59–61
Tablet with off-shelf transducers	5 ^{33,43,51,53,54}

Discussion

This systematic review focused on mHealth-validated screening applications that have been available for the paediatric population for the past decade (2014). The findings highlight that while mHealth-based paediatric hearing screeners have expanded their scope, they remain uneven across age groups and test paradigms. Most of the validated tools identified in this review were subjective hearing screeners available mainly for children above 3 years of age, while only a few objective OAE-based hearing screeners were identified, and no other mHealth-based objective hearing screeners were identified.

In this review, we were able to capture three mHealth-based objective OAE screeners such as OAEBuds,²⁸ Off the shelf OAE^27,29 and hearOAE³⁵ that have demonstrated higher sensitivity and comparable specificity when compared to conventional OAE screeners.^65,66 These validity measures indicate the technical feasibility of mobile integrated objective hearing screeners. However, two of the screeners were non-commercial prototypes, and only hearOAE³⁵ was commercially available at the time of this study. This suggests that, although analytical validity has been demonstrated, its translation into commercial screeners remains limited. The review did not identify any mHealth-based AABR screening tools. The AABRs are particularly used for high-risk neonates for identifying neurologic disorders.¹⁷ This absence highlights a critical gap in mHealth-based early detection. Although two mHealth systems, Otonova by otodynamics⁶⁷ that integrates both OAE and AABR and AScreen by neurosoft,⁶⁸ which has both TEOAE and DPOAE modules, though available globally their validation data were not identified in the peer-reviewed literature.

In contrast to objective hearing screeners, most of the validated tools identified were subjective hearing assessments. Many of these tools demonstrated acceptable to high sensitivity or strong agreement with diagnostic pure-tone audiometry, suggesting their utility similar to that of conventional hearing screening methods. However, the validity of these tests depends on the age of the child. Tools validated in older children (typically 4 years and above) demonstrated higher validity metrics than those in lower age groups. This is in line with the findings of Dawood et al.,⁶⁹ they screened children aged 3–10 years, compared the effectiveness of community health workers and school health nurses. It was reported that for each younger age, there was 0.214 times increase in referral results. This highlights the important limitation of subjective mHealth screening approaches for preschool-aged children and reinforces the need for innovative solutions for both subjective and objective screening modalities in younger populations.

In general, audiological testing requires soundproof rooms for testing, in which the maximum permissible ambient noise levels are in the range of 21–37 dB SPL,⁷⁰ to achieve good validity during hearing testing. In our review, we found that nearly 80% of the studies used quiet rooms in schools, communities and hospitals, and they achieved good screening validity. However, this review did not capture the screeners’ functionality in the presence of different room ambient noise conditions.

Only 16 out of 27 screeners were commercially available, and the others were prototypes or were discontinued. This is particularly important because validation alone does not equate to market readiness. Screeners also require country-specific regulatory approvals, long-term support and integration into health systems. Most screeners were developed and validated in high-income or upper middle-income countries. Eksteen et al.⁷¹ reported the cost analysis of hearing and vision mHealth screening program. The community health worker screened children of age 4–7 years using hearScreen for hearing screening and peek Acuity for vision screening. The study reported a program running cost of $5.63 per child. The cost of running a program for children younger than 4 years is unknown. Some of the mHealth tools use cloud data management system which requires an annual subscription. With the advantages of increased follow-ups, reporting, data traceability and advanced reporting,²² the upkeep/program costs for this needs to be studied.

From this study most mHealth tools were validated when used by community health workers or nurses,^12,57,72,73 to increase coverage and access to hearing screenings. While this model is successful in the number of children reached and followed up, it is also important to consider the workload of these community health workers, digital literacy and training aspects.⁷⁴

While mHealth-based OAE hearing screeners are emerging, validated mHealth-based ABR screeners were not identified through this systematic review. There's a lack of standardization of pass/referral criteria in hearing screening, which was observed in both objective and subjective screeners. This is required to improve comparability and program consistency among the screeners.

Despite the recent trend in AI and machine learning (ML), which can potentially enhance the screening accuracy and automation of referrals and follow-ups, there are no paediatric mHealth hearing screeners that have embedded this technology. This indicates that such technological innovations are yet to be translated for paediatric mHealth hearing screeners.

While the review highlights the advantages, there are some limitations. The included articles had more heterogeneity in the validation protocols and outcome measures. Hence, meta-analysis could not be done. Additionally, the review did not deeply examine the test metrics, test–retest reliability and device calibration procedures. Additionally, only English language papers were involved which might have excluded publications from non-English speaking context on their innovations.

All included studies required a quiet room or sound booth to conduct tests, but during mass screening procedures, ambient noise levels can be high. Therefore, it is essential to explore valid screening methods in noisy environments. Moreover, integrating AI into both objective (digital signal processing) and subjective assessments can help minimize human error while ensuring data privacy.

Conclusion

This systematic review demonstrates that mHealth paediatric hearing screeners can achieve validity comparable to conventional screening methods particularly for older childrens. However, the current evidence base is uneven with limited availability of objective-based mHealth hearing screeners.

Although several screeners achieved adequate validity, only a few were commercially available, with some available free of cost and as open source. Future research should emphasize the following: a standard validation protocol across age groups and contexts, the development of more objective hearing screeners and the adoption of the latest technological innovations, such as AI and machine learning, to increase the efficacy of these instruments.

Supplemental Material

sj-docx-1-dhj-10.1177_20552076261434816 - Supplemental material for A systematic review of mHealth-based paediatric hearing screening tools/devices

Supplemental material, sj-docx-1-dhj-10.1177_20552076261434816 for A systematic review of mHealth-based paediatric hearing screening tools/devices by Vishnu Saravana, Vidya Ramkumar and Arivudai Nambi Pitchaimuthu in DIGITAL HEALTH

Footnotes

Acknowledgements

Prof. Ashok Jhunjhunwala for his advisory role. Quillbot grammar checker and Paperpal have been used to curate the manuscript.

ORCID iDs

Vishnu Saravana

Vidya Ramkumar

Arivudai Nambi Pitchaimuthu

Ethical approval

The systematic review protocol was registered on the Prospective Register Of Systematic Reviews (PROSPERO) under the registration number CRD42024536576 (). There are no human participants in this study, hence informed consent is not required.

Funding

The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This work was supported by the Prime Minister Fellowship for Doctoral Research – ANRF, Shri N.P.V Ramasamy Udayar Research Fellowship, (grant number Batch 19, U022300983).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Data availability

All data analysed in this study is available in Scopus, PubMed and Google Scholar.

Supplemental material

Supplemental material for this article is available online.

References

WHO. Universal Newborn Screening Implementation Guidelines for Hearing Impairment, Eye abnormalities and newborn hyperbilirubinaemia at discharge , 2024. http://apps.who.int/bookorders .

Kanji

Khoza-Shangase

Moroe

. Newborn hearing screening protocols and their outcomes: A systematic review. Int J Pediatr Otorhinolaryngol 2018; 115: 104–109.

Mahomed-Asmail

Swanepoel

Eikelboom

. Hearing loss in urban South African school children (grade 1 to 3). Int J Pediatr Otorhinolaryngol 2016; 84: 27–31.

Chowdhury

Ravi

. Healthcare accessibility in developing countries: A global healthcare challenge. J Clin Biomed Res 2022; 4: 1–5.

WHO. MHealth : new horizons for health through mobile technologies. World Health Organization. https://www.afro.who.int/publications/mhealth-new-horizons-health-through-mobile-technologie (2011 , accessed 5 January 2026).

Khosravi

Azar

A systematic review of reviews on the advantages of mHealth utilization in mental health services: A viable option for large populations in low-resource settings. Glob Ment Health 2024; 11: e43. Epub ahead of print 4 April 2024.

Swanepoel

. Advancing equitable hearing care: Innovations in technology and service delivery. Folia Phoniatr Logop 2023; 75: 201–207.

Smartphone owners are now the global majority, New GSMA report reveals - Newsroom, https://www.gsma.com/newsroom/press-release/smartphone-owners-are-now-the-global-majority-new-gsma-report-reveals/ (accessed 17 January 2026).

Ministry of statistics and Programme Implementation, National Statistics Office. Comprehensive Modular Survey: Telecom. India (May 2025. https://www.mospi.gov.in/sites/default/files/publication_reports/CMST_report_m.pdf (accessed 11 September 2025).

10.

Bhatt

Liu

Gong

, et al. Emerging artificial intelligence–empowered mHealth: Scoping review. JMIR Mhealth Uhealth 2022; 10: e35053.

11.

Frisby

Eikelboom

Mahomed-Asmail

, et al. m-Health applications for hearing loss: A scoping review. Telemed J E Health 2022; 28: 1090–1099.

12.

Manus

van der Linde

Kuper

, et al. Community-based hearing and vision screening in schools in low-income communities using mobile health technologies. Lang Speech Hear Serv Sch 2021; 52: 568–580.

13.

Bush

Kaufman

McNulty

. Disparities in access to pediatric hearing health care. Curr Opin Otolaryngol Head Neck Surg 2017; 25: 359–364.

14.

Abbey

Douglas

TS.

Mobile phone applications to screen for hearing loss in low-and middle-income countries: a state-of-the-art review . 2021.

15.

Bright

Pallawela

. Validated smartphone-based apps for ear and hearing assessments: A review. JMIR Rehabil Assist Technol 2016; 3: 1–12.

16.

Chen

Lin

HYH

Wang

, et al. Diagnostic accuracy of smartphone-based audiometry for hearing loss detection: Meta-analysis. JMIR Mhealth Uhealth 2021; 9: e28378. Epub ahead of print 1 September 2021.

17.

The Joint Committee on Infant Hearing. Year 2019 position statement: Principles and guidelines for early hearing detection and intervention programs. J Early Hear Detect Intervent 2019; 4: 1–44.

18.

Dhar

Hall

. OAEs and Cochlear pathophysiology: Adults. Second. USA: Plural Publishing Inc., 2018.

19.

Northern

Downs

Hayes

. Hearing in children. 6th ed. Plural Publishing, https://books.google.com/books/about/Hearing_in_Children_Sixth_Edition.html?id=8W9yDgAAQBAJ (2014, accessed 27 January 2026).

20.

American Speech Language Hearing Association. Scope of practice in audiology ad hoc committee on the scope of practice in audiology about this document , 2018. www.asha.org/policy/ .

21.

Swanepoel

. Enhancing ear and hearing health access for children with technology and connectivity. Am J Audiol 2017; 26: 426–429.

22.

Bernstein

Besser

Maidment

, et al. Innovation in the context of audiology and in the context of the internet. Am J Audiol 2018; 27: 376–384.

23.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 2021; 372: 1–9.

24.

Schardt

Adams

Owens

, et al. Utilization of the PICO framework to improve searching PubMed for clinical questions. BMC Med Inform Decis Mak 2007; 7: 16.

25.

JBI Critical Appraisal Tools | JBI, https://jbi.global/critical-appraisal-tools (accessed 1 September 2025).

26.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 2021: n71.

27.

Ali

Chan

Meehan

, et al. An open-source smartphone otoacoustic emissions test for infants. Pediatrics 2025; 155: e2024068068. Epub ahead of print February 2025. 10.1542/peds.2024-068068.

28.

Chan

Glenn

Itani

, et al. Wireless earbuds for low-cost hearing screening. In: MobiSys 2023 - Proceedings of the 21st Annual International Conference on Mobile Systems, Applications and Services. Association for Computing Machinery, Inc, 2023, pp. 84–95.

29.

Chan

Ali

Najafi

, et al. An off-the-shelf otoacoustic-emission probe for hearing screening via a smartphone. Nat Biomed Eng 2022; 6: 1203–1213.

30.

Pereira

Pasko

Supinski

, et al.

Is there a clinical application for tablet-based automated audiometry in children?

Int J Pediatr Otorhinolaryngol 2018; 110: 87–92.

31.

Manayan

Ladd-Luthringshauser

Packer

, et al. Ambient noise limits efficacy of smartphone-based screening for hearing loss in children at risk. Am J Otolaryngol Head Neck Med Surg 2022; 43: 103223. Epub ahead of print 1 January 2022.

32.

Manganella

Stiles

Kawai

, et al. Validation of a portable hearing assessment tool: Agilis Health Mobile Audiogram. Int J Pediatr Otorhinolaryngol 2018; 113: 94–98.

33.

Kung

Kunda

Groff

, et al. Validation study of kids hearing game: A self-administered pediatric audiology application. Perm J 2021; 25: 79. Epub ahead of print 2021.

34.

Wiseman

Slotkin

Spratford

, et al. Validation of a tablet-based assessment of auditory sensitivity for researchers. Behav Res Methods 2023; 55: 2838–2852.

35.

Madzivhandila

le Roux

Biagio de Jager

Neonatal hearing screening using a smartphone-based otoacoustic emission device: A comparative study. Int J Pediatr Otorhinolaryngol 2024; 177: 1–8.

36.

Wolmarans

De Sousa

Frisby

, et al. Speech recognition in noise using binaural diotic and antiphasic digits-in-noise in children: Maturation and self-test validity. J Am Acad Audiol 2021; 32: 315–323.

37.

Louw

Swanepoel

Eikelboom Robert

, et al. Smartphone based hearing screening at primary health care clinics.

38.

Mahomed-Asmail

Swanepoel

Eikelboom

, et al. Clinical validity of hearScreen^TM smartphone hearing screening for school children. Ear Hear 2016; 37: e11–e17.

39.

Phanguphangu

Ross

. Clinical utility of smartphone-based audiometry for early hearing loss detection in HIV-positive children: A feasibility study. Afr J Prim Health Care Fam Med 2021; 13: 3077. Epub ahead of print 30 September 2021.

40.

De Swanepoel

Myburgh

Howe

, et al. Smartphone hearing screening with integrated quality control and data management. Int J Audiol 2014; 53: 841–849.

41.

Yalamanchali

Albert

Staecker

, et al. Evaluation of portable tablet-based audiometry in a South Indian population. Indian J Otolaryngol Head Neck Surg 2022; 74: 3592–3598.

42.

Pitchaimani

Rakesh

Bharathi

, et al. New age hearing testing at your fingertips: Shravana Mitra app. Indian J Otolaryngol Head Neck Surg 2024; 76: 322–328.

43.

Ramkumar

Joshi

Prabhakar

, et al. Development and beta validation of an mHealth-based hearing screener (SRESHT) for young children in resource-limited countries: Pilot validation study. JMIR Form Res 2025; 9: e43560. Epub ahead of print 2025. DOI: 10.2196/53460.

44.

Bowers

Graydon

Rance

. Evaluation of a game-based hearing screening program for identifying hearing loss in primary school-aged children. Int J Audiol 2023; 62: 512–520.

45.

Mealings

Harkus

Flesher

, et al. Detection of hearing problems in aboriginal and Torres strait islander children: A comparison between clinician-administered and self-administrated hearing tests. Int J Audiol 2020; 59: 455–463.

46.

Dillon

Mee

Moreno

, et al. Hearing tests are just child’s play: The sound scouts game for children entering school. Int J Audiol 2018; 57: 529–537.

47.

Bauer

Sales

Teixeira

, et al. Development and accuracy of a hearing screening application. Braz J Otorhinolaryngol 2021; 87: 711–717.

48.

Samelli

Rabelo

Sanches

SGG

, et al. Tablet-based tele-audiometry: Automated hearing screening for schoolchildren. J Telemed Telecare 2020; 26: 140–149.

49.

Corona

Ferrite

Bright

, et al. Validity of hearing screening using hearTest smartphone-based audiometry: Performance evaluation of different response modes. Int J Audiol 2020; 59: 666–673.

50.

Sudiro

Hasansulama

Digdayan Makerto

, et al. Screen-H website audiometry as a diagnostic tool for hearing screening in patients with hearing loss. F1000Res 2024; 13: 416.

51.

Rahim

Sunjaya

Hilmanto

, et al. Hearing screening alternative using a website-based application. J Audiol Otol 2023; 27: 123–127.

52.

Yimtae

Israsena

Thanawirattananit

, et al. A tablet-based mobile hearing screening system for preschoolers: Design and validation study. JMIR Mhealth Uhealth 2018; 6: e9560. Epub ahead of print 1 October 2018.

53.

Lü

, et al. A new hearing screening system for preschool children. Int J Pediatr Otorhinolaryngol 2014; 78: 290–295.

54.

Xiao

Zou

Gao

, et al. A novel tablet-based approach for hearing screening of the pediatric population, 516-patient study. Laryngoscope 2020; 130: 2245–2251.

55.

Van den Borre

Tufatulin

Zupan

, et al. A language-independent hearing screening self-test at school-entry. Sci Rep; 2024; 14: 2582. Epub ahead of print 1 December 2024.

56.

Van den Borre

Denys

Zupan

, et al. Language-independent hearing screening – increasing the feasibility of a hearing screening self-test at school-entry. Trends Hear 2022; 26: 1–15.

57.

Saunders

Bessen

Magro

, et al. Community health workers and mHealth systems for hearing screening in rural Nicaraguan schoolchildren. J Glob Health 2022; 12: 1–12.

58.

Sørensen

Adams

Pedersen

, et al. AMTASTM and user-operated smartphone research application audiometry—An evaluation study. PLoS One 2023; 18: 1–18.

59.

Chu

Cheng

Lai

, et al. A mobile phone–based approach for hearing screening of school-age children: Cross-sectional validation study. JMIR Mhealth Uhealth 2019; 7: 1–13.

60.

Durgut

Ekim

DIkici

, et al. Evaluation of hearing thresholds by using a mobile application in children with otitis media with effusion. Audiol Neurotol 2020; 25: 120–124.

61.

Sheikh Rashid

Dreschler

de Laat

JAPM

. Evaluation of an internet-based speech-in-noise screening test for school-age children. Int J Audiol 2017; 56: 967–975.

62.

Denys

Hofmann

Luts

, et al. School-age hearing screening based on speech-in-noise perception using the digit triplet test. Ear Hear 2018; 39: 1104–1115.

63.

Fordington

Brown

TH.

An evaluation of the hear glue ear mobile application for children aged 2–8 years old with otitis media with effusion. Digit Health 2020; 6: 1–16.

64.

Tananuchittikul

Yimtae

Chayaopas

, et al. App-based hearing screenings in preschool children with different types of headphones: Diagnostic study. JMIR Mhealth Uhealth 2023; 11: 1–9.

65.

Alarcón Avila

Saenz González

Vivas Gómez

, et al. Diagnostic performance and clinical follow-up of a universal newborn hearing screening program using OAE and ABR in a middle-income country. Int J Pediatr Otorhinolaryngol 2025; 197: 1–8.

66.

Heidari

Olyaee Manesh

Rajabi

The sensitivity and specificity of automated auditory brainstem response and otoacoustic emission in neonatal hearing screening: A systematic review. Auditory and vestibular research 2015; 24: 141–151.

67.

Otodynamics OtoNova - Fast ABR+OAE newborn and infant screener. https://www.otodynamics.info/product/otonova/ (accessed 14 January 2026).

68.

aScreen Neurosoft, https://neurosoft.com/en/catalog/audio/ascreen#overview (accessed 14 January 2026).

69.

Dawood

Mahomed Asmail

Louw

, et al. Mhealth hearing screening for children by non-specialist health workers in communities. Int J Audiol 2020; 60: S23–S29.

70.

Chung

Calibration matters: II. Measurement of ambient noise in test rooms/areas. J Commun Disord 2023; 101: 1–15.

71.

Eksteen

Launer

Kuper

, et al. Hearing and vision screening for preschool children using mobile technology, South Africa. Bull World Health Organ 2019; 97: 672–680.

72.

Yousuf Hussein

Swanepoel

Mahomed

, et al. Community-based hearing screening for young children using an mHealth service-delivery model. Glob Health Action 2018; 11: 1–8.

73.

Firoozbakht

Mahmoudian

Alaeddini

, et al. Community-based newborn hearing screening programme for early detection of permanent hearing loss in Iran: An eight-year cross-sectional study from 2005 to 2012. J Med Screen 2014; 21: 10–17.

74.

Meghana

RVNSP

Singh

Huda

. Challenges, barriers, and facilitators in implementing digital health initiatives for NP-NCD management in India’s public health system: A scoping review. Clin Epidemiol Glob Health 2025; 35: 102137.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.42 MB