Table 2 Comparison of tools for screening ear health and hearing
Tools for screening ear health | |||||
Author/year and study type | Sample | Study setting | Ref test | Comparator test | Key findings and effect measures |
Chianese et al, 2007 [48] Non-randomised comparative study | 786 healthy children aged 2-24 months | Primary care, US | Tympanometry and Spectral Grade Acoustic Reflectometry (SGAR) | Pneumatic otoscopy | - Tympanometry slightly more discerning than SGAR in predicting middle ear fluid: tympanometry AUC = 0.83 and SGAR AUC = 0.78 - 95%CI not reported |
Helenius et al, 2012 [49] Non-randomised comparative study | 515 children 0.5-3 years (mean 16 months) | Primary health care, Finland | Tympanometry | Pneumatic otoscopy | - All peaked tympanograms could be taken as healthy middle ears in primary care - Tympanometry unable to differentiate acute OM (AOM) from OM with effusion (OME) - For asymptomatic children: when middle ear aerated, tympanogram peaked in 97% of exams; when AOM diagnosed, tympanogram flat in 46% of exams; when persistent OME diagnosed, tympanogram flat in 71% of exams - Tympanometry unclear/not obtained: 24% - No analytical statistics reported |
Abbott et al, 2014 [50] Cross over study | 347 children aged 0.5-6 years | GPs in primary health care, Australia | Tympanometry and pneumatic otoscopy | Standard otoscopy | - After performing tympanometry or pneumatic otoscopy, GPs were three times more likely to amend diagnosis (χ 2 = 28.64, df 1, p < 0.001) and management plan (χ 2 = 9.24, df 1, p < 0.01) made on basis of otoscopy alone - GPs preferred tympanometry, but felt cost was a barrier to routine use |
Puhakka et al, 2014 [51] Non-randomised comparative study | 600 children aged 0.6 to 14 years | Study physicians in primary health care, Finland | Tympanometry and Spectral Grade Acoustic Reflectometry (SGAR) | Pneumatic otoscopy | - Good observed agreement (86%) between SGAR and tympanometry in children - Advantages of SGAR: low cost, portability, and no need for an airtight seal - SGAR sensitivity 53% (46-59), specificity 93% (92-94), positive predictive value 48% (41-53) and negative predictive value 94% (93-95) - Tympanometry sensitivity 56% (50-62), specificity 96% (95-96), positive predictive value 60% (53-66) and negative predictive value 95% (94-96) |
Alenezi et al, 2021 [52] Non-randomised comparative study | 157 children aged 0.5-15 years | ENTs, audiologists, trained assistants at community events, Australia | Video-otoscopy images | Standard otoscopy | - Video-otoscopy images produced significantly higher quality images than traditional otoscopy, across almost all domains rated (p < 0.05) - Image quality significant reduced with younger patient age (p < 0.03) |
Kleinman, K et al, 2021 [42] Randomised controlled trial | 197 children aged 0-21 years (48% aged 0-2 years, 32% aged 3-7 years) | Paediatric emergency department and primary care clinic, US | Smartphone video-otoscopy | Standard otoscopy | - Accuracy of ear examination findings using smartphone otoscope improved by 11.2% (95% CI: 1.5, 21.8%, p = 0.033) relative to traditional otoscopy, to 74.8% (95% CI: 67.3, 82.1%) |
Tools for screening hearing | |||||
Author/year and study type | Sample | Study setting | Ref test | Comparator test | Key findings |
Newton et al. 2001 [46] Non-randomised comparative study | 757 children aged 2.2-7.5 years. Median age 5.4 years. | Community nurses in nursery schools and child health clinics, Kenya | 8-question parent/ caregiver questionnaire exploring behavioural responses to sound and communication ability designed to detect bilateral hearing loss > 40 dB HL. | ENT and audiological evaluation by ENT Clinical Officers | - 100% sensitive for bilateral hearing loss of 40 dB HL and greater and 75% specific when compared against audiometry thresholds and ENT ear observations - Negative predictive value was 100%, but positive predictive value was low, at 6.75%. - No confidence intervals reported - Authors concluded that the questionnaire, administered by healthcare workers, could be usefully applied in primary healthcare for detecting hearing impairment at the pre-school stage |
Mahomed-Asmail et al. 2016 [45] Non-randomised comparative study | 1070 children aged 5-12 years; average age 8 years. | Primary schools, South Africa | hearScreen smartphone screening app and conventional screening audiometry | Diagnostic audiometry | - No significant difference in performance - Smartphone screener and conventional screening demonstrated equivalent sensitivity (75%) and similar specificity (98.5% and 97% respectively) - Positive and negative predictive values 52.9% and 99.4% for smartphone screener, and 36.7% and 99.4% for conventional hearing screening. - No confidence intervals reported |
Ramkumar et al 2018 [47] Non-randomised comparative study | 119 children (43) and young infants (76) aged 0-5 years | Trained village health workers, community setting, India | Distortion Product Otoacoustic Emissions (DPOAE) | Tele-Auditory Brainstem Response testing | - The study found acceptable validity: sensitivity of DPOAE screening was 75% (CI: 69-81) and specificity, 91% (CI: 87-95) - Negative and positive predictive values were 99% (CI: 98-100) and 27% (CI: 21-33), respectively |
Mealings et al, 2020 [44] Non-randomised comparative study | 297 Aboriginal and Torres Strait Islander children aged 4–14 years | Primary schools, Australia | Sound Scouts game-based hearing test app for smartphones and tablets | Pure tone audiometry, Listening in Spatialised Noise – Sentences high-cue condition | - Sensitivity of Sound Scouts for average hearing loss of >20 dB HL was 41% and specificity was 89%; and for average hearing loss >30 dB HL, sensitivity at 88% and specificity at 88% - Consistent pass/fail results on Sound Scouts speech-in-noise measure and Listening in Spatialised Noise – Sentences test high-cue condition were found for 73% of children |
Orzan et al. 2021 [53] Non-randomised comparative study | 309 children aged 1-36 months | Oto-rhino-laryngology and audiology unit of a medical institute, Italy | Parental assessment of auditory skills using the Questionnaire on Hearing and Communication Abilities (QUAC) | Audiological evaluation of children at a secondary care institute | - Parents reported a decrease in auditory skills for children with sensorineural hearing loss (Χ2(2)=14.4, p=0.003), with increased concern expressed in 59% compared with 24% in normally hearing children - Positive predictive value was 0.78, but with low sensitivity (0.39) - No confidence intervals reported - Conclusion: parents have capacity to recognise non-typical auditory behaviours; an auditory abilities checklist can complement existing primary healthcare screening procedures |