Hearing loss occurs when the sound transmission from the outer ear to the brain, is impaired. ¹ In children, it is associated with impairments in expressive and receptive language skills, challenges in social interactions, stress management issues, increased frustration, and lowered self-esteem. ^{2 – 6} In adults, hearing loss affects communication, daily activities, work, social life, and self-confidence. Furthermore, it is linked to occupational restrictions, leading to economic issues. ^{2 – 6}

Early detection of hearing loss is essential for timely and accurate intervention. ^{5 , 7} Pure-tone audiometry is considered the gold standard for diagnosing hearing loss; however, limitations in infrastructure, availability of trained personnel, and the requirement for soundproofed environments can delay hearing loss evaluations, causing gaps in hearing screening in rural or remote areas. ^{5 , 7} An alternative approach is hearing screening using Screen-H ^, which offers a rapid, painless, non-invasive solution that is accessible through mobile phones, tablets, and computers, provided there is a signal and headphones for sound stimulation. ^{7 – 11}

This study aimed to evaluate the agreement between Screen-H and pure-tone audiometry.

The study population comprised patients with hearing loss seeking treatment at the Audiology Clinic of Otorhinolaryngology, Head and Neck Surgery, at Dr. Hasan Sadikin General Hospital, Bandung. This analytical observational research employed a cross-sectional design and a diagnostic test to evaluate the performance of Screen-H. The diagnostic test assessed multiple measures, including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (PLR), negative likelihood ratio (NLR), and accuracy. The reliability of the study was assessed using the Kappa Index. Data processing and analysis were conducted using Microsoft Excel 365 https://www.microsoft.com/id-id/microsoft-365/free-office-online-for-the-web. The inclusion criteria for this study were individuals aged over 10 years who demonstrated a cooperative demeanor and havDatae agreed to research consent. The exclusion criteria included the inability to undergo pure tone audiometry, craniofacial anomalies, and congenital abnormalities. This study received approval from the ethics committee of Padjadjaran University, under reference number 255/UN6.KEP/EC/2023 which was published on 28 February2023, and was conducted from May to July 2023.

Pure tone audiometry was performed in a soundproof room. Screen-H hearing tests were administered in a room with a noise level ≤ 40 dB using Sennheiser HD 201 headphones, chosen for their noise reduction capability of 15-20 dB HL, a stimulus output of up to 108 dB HL, and a frequency range of 21-18000 Hz. The calibration process for Screen-H relied on a sound reference approach, measuring sound intensity and subsequently record it to establish a reference point. Post-calibration, it was determined that the sound intensity emitted by Screen-H aligned with that of pure tone audiometry. The typical ambient sound level in the test room was measured at 40 dB SPL. ¹² During the hearing screening, the initial stimulus intensity commonly ranged from 30-40 dB HL, with the recommended initial intensity for Screen-H set at 35 dB HL. ¹²

Hearing tests were performed using Screen-H and pure tone audiometry (Resonance, EN60645-1/-2, New Holland) as the gold standard. Screen-H has been patented and we as researchers have permission to use the application and Software available from: screenout.id. We use pure tone audiometry (Resonance, EN60645-1/-2, New Holland) which has permission to use from the local health department. All subjects underwent examinations first with Screen-H and then with pure tone audiometry. During the Screen-H test, subjects were exposed to auditory stimuli through headphones at specific frequencies of 500, 1,000, 2,000, 4,000, and 8,000 Hz, with intensity levels set at 35, 55, and 75 dB. ¹³

A total of 321 participants met the inclusion and exclusion criteria, comprising 280 individuals with hearing impairment and 41 with normal hearing.

The most prevalent age range for hearing loss complaints was between 26 and 35 years. Table 1 presents the prevalence rates of hearing loss among males and females, showing a rate of 53% for males and 47% for females. Sensorineural hearing loss was identified as the most common type, with chronic suppurative otitis media being the leading cause.

Tables 2 and 3 present results indicating that Screen-H closely adheres to pure tone audiometry in both the right and left ears, as evidenced by high Kappa values.

Table 4 shows that Screen-H exhibits strong sensitivity, very high specificity, very high PPV and NPV in the right ear, with sufficient NPV, high accuracy, high PLR, high NLR in the left ear. The Kappa value suggests greater suitability for the right ear compared to the left.

Hearing loss affects individuals across all ages. ^{4 , 14} Our study included patients ranging from 11 to 83 years, with the most prevalent age group being 26 to 35 years. There is higher prevalence of hearing loss in men compared to women, which can be attributed to several risk factors and comorbidities. ¹⁵

The current study revealed that sensorineural hearing loss was the most common type, accounting for 40.5% of cases, followed by conductive hearing loss at 26%. Various factors contributed to these types of hearing loss, including chronic suppurative otitis media, presbycusis, ototoxicity, noise-induced hearing loss, sudden sensorineural deafness, congenital hearing loss, and Meniere's disease. Additionally, cerumen presence was a contributor to conductive hearing loss.

Chronic suppurative otitis media has been linked with various types of hearing loss, including conductive, sensorineural, and mixed types. ¹⁶ Particularly, sensorineural hearing loss can develop when chronic otitis media persists for more than 14 days, leading to increased levels of CD45+ macrophages, which act as inflammatory mediators in the cochlea. These macrophages are closely associated with the damage to the external hair cells. ¹⁶ Prolonged infection and inflammation beyond 14 days can lead to irreversible auditory function impairment. primarily affecting Based on Xia’s research, the damage primarily affects the outer hair cells while sparing the inner hair cells. ^{11 , 16 , 17} Conductive hearing loss, on the other hand, results from impaired sound transmission from the outer to the inner ear. In this study, factors such as chronic otitis media, cerumen prop, acute otitis media, external acoustic canal stenosis, tumors, and otosclerosis accounted for the 26% incidence of conductive hearing loss. ¹⁸

Screen-H has greater specificity compared to sensitivity, as the test is more effective at accurately identifying individuals without hearing loss than those with hearing loss. The accuracy of the test in both the right and left ears demonstrated robust values, indicating that the audiometry testing conducted on the Screen-H possesses statistically significant accuracy. The Kappa value is strong for the right ear and satisfactory for the left ear.

Other studies have reported sensitivity and specificity rates of 82% and 95%, respectively, which are comparable to the findings of our current study. ¹⁹ Another study revealed a sensitivity of 93%, specificity of 88%, a positive predictive value of 87%, and a negative predictive value of 94%. ²⁰ These findings align with our investigation, indicating that the Screen-H assessment has adequate statistical significance in accurately identifying individuals with either hearing loss or normal hearing. The impact of environmental noise on website audiometry examinations is major, especially when tests are carried out at intensities >40 dB HL. ^{21 , 22} In this study, Screen-H examinations were performed in areas where the environmental noise was ≤40 dB HL, and headphones with a noise-canceling capability of around 15-20 dB HL were used. This setup ensured that the 35 dB stimulus was effectively received.

In addition, the earphones can affect the results of audiological assessments and whether the earphones are positioned by trained examiners or by examinees can affect audiological judgments. ¹¹ The negative effects of background noise may further support that examinations performed in soundproof booths have better diagnostic accuracy. ¹⁹ The influence of ambient noise, which results in incorrect smartphone audiometry test results, leading to the use of soundproof booths can improve the diagnostic accuracy of smartphone-based hearing tests. ¹¹

Pure tone audiometry, known for its high cost and limited accessibility, is typically available only in larger healthcare facilities. This presents a challenge for many individuals in Indonesia, especially considering the country’s archipelagic geography, which further complicates access to hearing examination services. Therefore, there is an urgent need for a hearing loss detection method that is accessible to all Indonesians, cost-effective, available in the Indonesian language, and does not require the assistance of professional personnel.

This study has limitations, notably the use of Sennheiser HD 201 headphones, which are not standard audiometry equipment. This factor could potentially influence the study results.

Screen-H demonstrated excellent capabilities in hearing loss screening due to its speed, ease of use, non-invasive nature, high sensitivity (80.8%-82.8%) and specificity (93.9%-(94.1%), and cost-effectiveness. It can effectively be used as a screening tool for hearing loss in environments with noise levels ≤ 40 dB, starting with a stimulus of 35 dB.

This research has been approved by the Ethics Committee of Padjadjaran University with number 255/UN6.KEP/EC/2023 which was published on February 28, 2023 and was conducted from May to July 2023. All participants were informed about subject of the study and signed written consent.

MS performed examination to the subject, analysis, and interpretation of the patient’s data, also was a major contributor in writing the manuscript. All authors had full access to all data in the study ad had final responsibility for the decision on submit for publication.