|Year : 2016 | Volume
| Issue : 2 | Page : 72-77
Extending otology services to the rural community: Use of smartphone for hearing screening
Abubakar Danjuma Salisu
Department of Otolaryngology, Bayero University Kano, Aminu Kano Teaching Hospital, Kano, Nigeria
|Date of Web Publication||1-Aug-2016|
Abubakar Danjuma Salisu
Department of Otolaryngology, Bayero University Kano, Aminu Kano Teaching Hospital, Kano
Source of Support: None, Conflict of Interest: None
Background: Hearing screening services scarcely extend to rural communities due to limited resources. Active intervention to prevent hearing loss may, therefore, be ineffective. Hearing screening with the smartphone may extend screening services to rural communities. This paper aims to assess the reliability of using the smartphone for hearing screening by comparing results obtained with those of a standard calibrated audiometer. Methodology: A hospital-based cross-sectional study of 60 participants consisting of 36 males and 24 females, age ranged from 16 years to 68 years (mean = 33.4, standard deviation = 12.7). Using the Android application 'Hearing Test Pro™' installed on Samsung Galaxy Note 3 smartphone, hearing threshold across 250 Hz–8000 Hz frequencies of participants was determined in a quiet clinic room. This was repeated using a calibrated Welch/Allyn TM 262 auto tympanometer/audiometer. Hearing thresholds at low, high and speech range frequencies were summarised with threshold >40 dB considered as 'Failed screening'. The results obtained from the two methods were compared. Results: There was complete agreement between the hearing threshold results obtained by the two methods at high frequencies with smartphone sensitivity 96% and specificity 100%. There was a moderate agreement between the two methods at speech range frequencies with smartphone sensitivity 86% and specificity 99%. The least agreement between the two methods was in the low frequencies. Conclusion: The smartphone application 'Hearing test Pro' is quite accurate, especially in detecting high frequency hearing loss and can be used effectively in hearing screening.
Keywords: Hearing, screening, smartphone
|How to cite this article:|
Salisu AD. Extending otology services to the rural community: Use of smartphone for hearing screening. Niger J Basic Clin Sci 2016;13:72-7
|How to cite this URL:|
Salisu AD. Extending otology services to the rural community: Use of smartphone for hearing screening. Niger J Basic Clin Sci [serial online] 2016 [cited 2019 Jul 23];13:72-7. Available from: http://www.njbcs.net/text.asp?2016/13/2/72/187359
| Introduction|| |
The consequence of hearing loss includes problems with communication, language development and educational problems. Early detection through screening prevents disabilities from ensuing. Conventional hearing screening facility to detect early hearing loss is not generally accessible to most rural communities, precluding early intervention.
The prevalence of hearing loss varies with age with increasing age being a risk factor. Other risk factors include ototoxicity, smoking, hypertension or diabetes.,, Noise exposure from earphone use has become frequent.
Screening could be effected using questionnaires, clinic tests such as whispered voice, finger rub or the more objective screening audiometer. A hearing screening using the audiometer may require a trained personnel in addition to the equipment being expensive. Audiometric screening could also be automated with the subject self-administering the test. Hearing screening could also be done over the telephone utilising the digit triplet test., While virtually all hearing screening methods have been validated to be accurate, they have the disadvantage of not reaching the rural communities satisfactorily. However, with increasing distribution of mobile phones worldwide including rural communities, telehealth appears to have the best potential to reach these underserved communities in the near and foreseeable future.
Mobile phone technology has been used in several health screening instances. This includes cardiovascular screening amongst others., The use of smartphones to screen for hearing loss has been studied by many researchers. The smartphone applications uHear™, HearScreen™ on iPhone and iPods have been found to be as accurate as conventional screening audiometer in detecting hearing impairment.,,,,
Android phones, however, tend to be cheaper and more available in this setting. Several applications have been developed for hearing screening on the Android platform, but none to the best of our knowledge has been validated in this environment.
This study used the android application 'Hearing Test Pro™' developed by E-audiologia.pl on Google play installed on the Samsung Galaxy Note 3 phone to screen for hearing impairment with the aim of determining its reliability for use as a screening tool in the rural setting by comparing the results obtained with those of a screening audiometer.
| Methodology|| |
This was a hospital-based cross-sectional study carried out over a 6 weeks period involving volunteer staff and patients who attended out-patient clinic and ENT clinic of the hospital.
Sixty participants were selected randomly among hospital staff and hospital patients presenting with non-otologic complaints, anyone not consenting or too unwell to participate was excluded from the study. Due consent was obtained from participants and institutional approval was obtained from the ethical review committee. Instruments for this research included a questionnaire, Samsung Galaxy Note 3 phone (SM-N900), running Android version 5.0 with 'Hearing Test Pro' Android application installed and a screening audiometer. Each participant filled out or answered questions on the questionnaire, then proceeded to have an otoscopic examination, then operator administered a hearing test using Samsung Galaxy Note 3 with the bundled earphone of the phone. Masking of the non-test ear was automatically done by the phone app for the test at 45 dB and above. A second test was then carried out using conventional audiometer (Welch/Alynn TM 262 Auto Tympanometer/Audiometer, current calibration) by a trained audiologist/audiology assistant. A smartphone re-test of ten participants selected randomly was then carried out to verify test-retest reliability. All test was carried out in a quiet consulting room (sound pressure level range = 28 dB–46 dB).
The participant's hearing threshold for each ear over 250–8000 Hz frequencies using the two methods was recorded. Three pure tone averages (PTAs) were obtained for each participant. These included PTA in lower frequencies (250, 500, 1000 Hz), in higher frequencies (2000 Hz, 4000 Hz, 6000 Hz, 8000 Hz) and routine PTA (500, 1000, 2000 Hz, 4000 Hz). Average hearing threshold >40 dB was considered as 'FAIL', suggestive of hearing impairment. Data obtained was input into MedCalc Statistical Software version 16.1 (MedCalc Software BVBA, Ostend, Belgium; https://www.medcalc.org; 2016). Data were summarised in tables using frequencies. Paired sample t-test was used to compare the means of the hearing threshold obtained by the two methods. The level of statistical significance was set at P < 0.05.
| Results|| |
There were 60 participants representing 120 ears, consisting of 36 males and 24 females, age ranged from 16 years to 68 years (mean = 33.4 years, standard deviation [SD] =12.7).
Screening with smartphone and audiometer at higher frequencies (2000 Hz, 4000 Hz, 6000 Hz, 8000 Hz)
At higher frequencies, smartphone picked out 25 of 26 ears that have a hearing impairment (PTA > 40 dB) [Table 1].
|Table 1: Frequency table showing average hearing threshold (dB) of 120 ears screened at higher frequencies|
Click here to view
At the higher frequencies, the smartphone has a sensitivity of 96% and specificity of 100% for detecting hearing impairment (threshold >40 dB).
At the higher frequencies, paired sample t-test revealed hearing thresholds obtained by smartphone (mean = 1.65, SD = 0.80) were statistically the same with those obtained by audiometer (mean = 1.65, SD = 0.81) t = −0.33, P = 0.74.
Screening with smartphone and audiometer at lower frequencies (250 Hz, 500 Hz, 1000 Hz)
At lower frequencies, smartphone picked out 23 out of 33 ears that have a hearing impairment [Table 2]. Smartphone additionally wrongly diagnosed one ear as having a hearing impairment.
|Table 2: Frequency table showing average hearing threshold (dB) of 120 ears screened at lower frequencies|
Click here to view
At lower frequencies, smartphone has a sensitivity of 70% and specificity of 99% for detecting hearing impairment (threshold >40 dB).
At the lower frequencies, paired sample t-test revealed the hearing threshold obtained by the smartphone (mean = 1.73, SD = 0.77) was significantly different from thresholds obtained by audiometer (mean = 1.92, SD = 0.79) t (119) =4.44, P < 0.0001.
Screening using threshold averages at speech frequencies 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz
At speech frequencies, smartphone picked 25 of 27 ears that have a hearing impairment [Table 3] but additionally wrongly diagnosed one ear as having a hearing impairment.
|Table 3: Frequency table showing average hearing threshold (dB) of 120 ears screened at speech frequencies|
Click here to view
At speech frequencies, the smartphone has a sensitivity of 93% and specificity 99% for detecting hearing impairment (threshold >40 dB).
At PTA of 500 Hz, 1000 Hz, 2000 Hz, 4000 Hz, paired sample t-test revealed hearing threshold obtained by smartphone (mean = 1.71, SD = 0.80) was significantly different from those obtained by an audiometer (mean = 1.8, SD = 0.78). t = −2.55, P < 0.01.
Screening using the questionnaire
Of 22 participants suspected to have hearing impairment based on responses to questions of hearing difficulty and tinnitus in the questionnaire, 21 proved true with significant hearing impairment on screening by audiometer [Table 4]. The questionnaire, however, missed 8 (28%) participants who failed the audiometer screening. Questionnaire showed a sensitivity of 72% for detecting hearing impairment.
|Table 4: Results of responses in questionnaire compared with standard audiometry screening results|
Click here to view
Smartphone test-retest reliability (ten participants)
Of the 20 ears retested and 20 PTA averages obtained, 18 were consistent (intraclass correlation coefficient - 0.96). In all 20 ears retested, results of each ear tested were identical to the result of the initial test for the ear as either 'PASS' or 'FAIL'.
| Discussion|| |
Hearing screening programmes are particularly important for the 'at risk' individuals to allow for timely intervention to prevent the development of hearing loss and its attendant consequences. This group includes but are not limited to the elderly who are at risk of degenerative changes of ageing and the young adults at risk from exposure to noise through misuse of headphones and earphones connected to smartphones. In the US, a study found an increase of 75% between 1990 and 2005 in the number of individuals listening to music through headphones and earphones. Typically, users of earphones and headphones listen to high volumes and over prolonged periods effectively putting hearing at risk. WHO estimates that 1.1 billion young people worldwide are at risk of hearing loss from unsafe use of personal audio devices. While the ownership and use of the smartphone poses a risk to hearing, it's installed applications do have a potential for screening of hearing impairment.
In comparing the readings obtained by the audiometer and the smartphone, this study found that at the higher frequencies of 2000 Hz and above, there was complete agreement between readings obtained by the smartphone and those obtained by the audiometer. This is similar to the findings of Peer and Fagan. At high frequencies, the Android Hearing test Pro™ showed a sensitivity of 96% and a specificity of 100% for detecting hearing impairment (PTA >40 dB), this is comparable to 96%-100% sensitivity of iPhone uHear™ application noted by other researchers.,,
At lower frequencies below 2000 Hz however, the agreement between the readings by the two test techniques was less pronounced. When normal hearing was present (threshold <25 dB), smartphone tended to give readings of better hearing for the subjects compared to the audiometer, however when significant hearing impairment (threshold >40 dB) was present, the difference between readings of the smartphone and audiometer approximate each other, but smartphone reading was consistently in favour of better hearing. This contrasted with the findings of Abu-Ghanem et al. where the study observed that smartphone (uHear™ for iPhone) results showed poorer hearing when compared with audiometer results.
Background noise, use of quiet test environment instead of a soundproof booth, type and fitting of earphone/headphone as well as calibration issues may be responsible for these differences. In this study, the smartphone missed 27% of cases with significant hearing impairment at the lower frequencies, it showed a sensitivity of 70%. Other studies have also observed this difference in readings between the 2 techniques at lower frequencies and concluded that the smartphone had a moderate accuracy of detecting hearing loss in these frequencies.
The practical application of the smartphone as a screening tool for hearing in the field should not require the use of soundproof booths, and in keeping with this, all hearing test in this study were carried out in a quiet environment with background noise not completely eliminated. Although the test environment fairly satisfies the American National Standard Institute ANSI S 3.1-1991 specification for maximum permissible ambient noise, nevertheless, it is believed that the test result obtained, especially at the lower frequencies may have been affected by background noise. Since the noise level in the test environment alters from moment to moment, it is possible that the higher permissible noise level for high frequencies (46 dB SPL in the test environment in this study) may have been present at the material moment of obtaining low-frequency thresholds, resulting in inaccurate threshold estimation at lower frequencies. The mobile phone itself can be used to assess the level of background noise just before a hearing screening test is carried out. This may improve the accuracy of the screening test at low frequencies. Some researchers using the smartphone found no significant difference however in results obtained in a quiet environment and those obtained in a soundproof booth. Currently, active research is ongoing introducing the technology of active noise reduction into audiometry headphones or earphones which will allow for more accurate measurements of hearing in the field despite background noise.,
The type of ear fitting used in carrying out the test may also have affected the result. While the bundled earphone of the smartphone fitted into the ear canal producing a good occlusal effect and reduced background noise in this study, the supra-aural headphones bundled with the audiometer was judged not to be that effective in reducing background noise. In a study on effects of background noise on earphone thresholds, the authors observed that of 3 types of earphone studied, the supra-aural headphones were least effective in the prevention of background noise.
Calibration issues may also affect the results obtained. While the audiometer is routinely calibrated, the developers of the smartphone application do not directly calibrate the earphones bundled with the phones but provide an automatic download of the calibration coefficients for the specific model of the mobile device with its bundled earphones, the implication of this is that replacing/interchanging the earphones (a constant phenomenon amongst phone users) means a re-calibration is required before results can be valid.
The differences observed between readings of the two test techniques in the lower frequencies also resulted in differences when PTAs were obtained in the speech range that excludes 4000 Hz (500 Hz, 1000 Hz, 2000 Hz). It was, however, observed that the addition of hearing threshold at 4000 Hz frequency in the PTA estimation improved the sensitivity of the smartphone in detecting hearing impairment from 70% to 93%, with a specificity of 99%. As a result of this, it is advisable to use 4 frequencies of 500 Hz, 1000 Hz, 2000 Hz and 4000 Hz in obtaining the PTA averages in the speech range when screening with the smartphone. A study on school children using the hearScreen™ smartphone application reported a 75.0% sensitivity and a 98.5% specificity.
The questionnaire used in this study detected 21 (72%) cases of hearing impairment. This was similarly observed in a study on the elderly. A study on children population, however, reported questionnaire to be 100% sensitive in detecting hearing impairment. It is likely that parents'/caregivers give more accurate information regarding hearing status than afflicted adult individuals who self-report hearing loss as the case in this study. However, another study found perfect agreement between self-reported hearing difficulty with measured hearing loss in adults.
In 20 ears randomly selected for re-test using the smartphone, this study observed a high test-retest reliability (intraclass correlation coefficient = 0.96) which is similar to finding by Kam et al. This smartphone's advantage of self-test and re-test coupled with its portability and availability in homes makes it a feasible hearing screening tool for resource-limited settings.
| Conclusion|| |
The smartphone android application 'hearing test Pro™' is quite accurate in detecting, especially high-frequency hearing impairment, and moderately accurate for lower frequencies. When coupled with a properly structured questionnaire, the smartphone can be used effectively in hearing screening programmes in rural Nigeria. It allows any cadre of health personnel to conduct hearing screening of a large segment of the population that will otherwise not be covered by routine screening programmes. Further studies are however necessary to validate these findings on low-end Android smartphones and a multitude of other smartphone brands.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Sogebi OA. Assessment of the risk factors for hearing loss in adult Nigerian population. Niger Med J 2013;54:244-9.
Brant LJ, Gordon-Salant S, Pearson JD, Klein LL, Morrell CH, Metter EJ, et al.
Risk factors related to age-associated hearing loss in the speech frequencies. J Am Acad Audiol 1996;7:152-60.
Bainbridge KE, Hoffman HJ, Cowie CC. Risk factors for hearing impairment among U.S. adults with diabetes: National Health and Nutrition Examination Survey 1999-2004. Diabetes Care 2011;34:1540-5.
Vogel I, Brug J, Hosli EJ, van der Ploeg CP, Raat H. MP3 players and hearing loss: Adolescents' perceptions of loud music and hearing conservation. J Pediatr 2008;152:400-4.
Mahomed F, Swanepoel de W, Eikelboom RH, Soer M. Validity of automated threshold audiometry: A systematic review and meta-analysis. Ear Hear 2013;34:745-52.
Watson CS, Kidd GR, Miller JD, Smits C, Humes LE. Telephone screening tests for functionally impaired hearing: Current use in seven countries and development of a US version. J Am Acad Audiol 2012;23:757-67.
Jansen S, Luts H, Dejonckere P, van Wieringen A, Wouters J. Efficient hearing screening in noise-exposed listeners using the digit triplet test. Ear Hear 2013;34:773-8.
Swanepoel de W, Clark JL, Koekemoer D, Hall JW 3rd
, Krumm M, Ferrari DV, et al.
Telehealth in audiology: The need and potential to reach underserved communities. Int J Audiol 2010;49:195-202.
Surka S, Edirippulige S, Steyn K, Gaziano T, Puoane T, Levitt N. Evaluating the use of mobile phone technology to enhance cardiovascular disease screening by community health workers. Int J Med Inform 2014;83:648-54.
Matousek A, Paik K, Winkler E, Denike J, Addington SR, Exe C, et al.
Community health workers and smartphones for the detection of surgical site infections in rural Haiti: A pilot study. Lancet 2015;385 Suppl 2:S47.
Peer S, Fagan JJ. Hearing loss in the developing world: Evaluating the iPhone mobile device as a screening tool. S Afr Med J 2015;105:35-9.
Szudek J, Ostevik A, Dziegielewski P, Robinson-Anagor J, Gomaa N, Hodgetts B, et al.
Can Uhear me now? Validation of an iPod-based hearing loss screening test. J Otolaryngol Head Neck Surg 2012;41 Suppl 1:S78-84.
Handzel O, Ben-Ari O, Damian D, Priel MM, Cohen J, Himmelfarb M. Smartphone-based hearing test as an aid in the initial evaluation of unilateral sudden sensorineural hearing loss. Audiol Neurootol 2013;18:201-7.
Swanepoel de W, Myburgh HC, Howe DM, Mahomed F, Eikelboom RH. Smartphone hearing screening with integrated quality control and data management. Int J Audiol 2014;53:841-9.
Mahomed-Asmail F, Swanepoel de W, Eikelboom RH, Myburgh HC, Hall J 3rd
. Clinical validity of hearScreen™ smartphone hearing screening for school children. Ear Hear 2016;37:e11-7.
Henderson E, Testa MA, Hartnick C. Prevalence of noise-induced hearing-threshold shifts and hearing loss among US youths. Pediatrics 2011;127:e39-46.
Abu-Ghanem S, Handzel O, Ness L, Ben-Artzi-Blima M, Fait-Ghelbendorf K, Himmelfarb M, et al.
Smartphone-based audiometric test for screening hearing loss in the elderly. Eur Arch Otorhinolaryngol 2016;273:333-9.
Ibekwe TS, Folorunsho DO, Dahilo EA, Gbujie IO, Nwegbu MM, Nwaorgu OG, et al.
Evaluation of mobile smartphones app as a screening tool for environmental noise monitoring. J Occup Environ Hyg 2016;13:D31-6.
Sandström J, Swanepoel de W, Carel Myburgh H, Laurent C. Smartphone threshold audiometry in underserved primary health-care contexts. Int J Audiol 2016;55:232-8.
Na Y, Joo HS, Yang H, Kang S, Hong SH, Woo J. Smartphone-based hearing screening in noisy environments. Sensors (Basel) 2014;14:10346-60.
Lo AH, McPherson B. Hearing screening for school children: Utility of noise-cancelling headphones. BMC Ear Nose Throat Disord 2013;13:6.
Frank T, Williams DL. Effects of background noise on earphone thresholds. J Am Acad Audiol 1993;4:201-12.
Newton VE, Macharia I, Mugwe P, Ototo B, Kan SW. Evaluation of the use of a questionnaire to detect hearing loss in Kenyan pre-school children. Int J Pediatr Otorhinolaryngol 2001;57:229-34.
Ramkissoon I, Cole M. Self-reported hearing difficulty versus audiometric screening in younger and older smokers and nonsmokers. J Clin Med Res 2011;3:183-90.
Kam AC, Sung JK, Lee T, Wong TK, van Hasselt A. Clinical evaluation of a computerized self-administered hearing test. Int J Audiol 2012;51:606-10.
[Table 1], [Table 2], [Table 3], [Table 4]