Middle ear measures as predictors of hearing loss in Australian Aboriginal schoolchildren

Pediatric International Journal

of Pediatric Otorhinolaryngology 30 (1994) 15-27

Middle ear measures as predictors of hearing loss in Australian Aboriginal schoolchildren Terry G. Nienhuys *, Judith

B. Boswell, Fred B. McConnel

Menzies School o/ Hrulth Resrurch Dunvm. Auslrcrli~r

(Received

26 July 1993: revision

received

22 November

1993: accepted

28 November

1993)

Abstract This study sought to establish the value of tympanometry and otoscopy for predicting significant conductive hearing loss in remote-area Aboriginal children, and also to measure the range of hearing loss which can be expected with middle ear disease, with or without a tympanic membrane (TM) perforation. A field study is reported of 255 Aboriginal children aged up to 16 years who were examined with pneumotoscopy and tympanometry and whose hearing was tested under controlled acoustic conditions. Results showed that pneumatic otoscopy for detection of middle ear effusion and identification of perforations resulted in the best rate of prediction of significant conductive hearing loss. Furthermore, the hearing of children with perforated TMs (mean pure-tone average 30.0 dB; S.D. 11.I) was significantly worse than those in which tympanometry suggested middle ear effusion (mean pure-tone average 20.3 dB: S.D. 9.6), and both differed significantly from ears showing normal tympanograms (mean pure-tone average 11.2 dB; SD. 5.9). Implications for community-based hearing screening and classroom management of affected children are discussed. Key words: Otitis media; Screening;

Tympanometry;

Otoscopy;

Middle ear; Aboriginal

1. Introduction Although it is well established that otitis media (OM) and associated conductive hearing loss (CHL) are highly prevalent among Australian Aboriginal children. little is known about reliable ways of detecting these in remote Aboriginal health clinics and schools or the extent of hearing impairment associated with different OM condi* Corresponding

author,

PO Box 41096, Casuarina,

0 1994 Elsevier Science Ireland 0165-5876194/$07.00 SSDI 0165-5876(93)01011-8

NT. 081 I. Australia.

Ltd. All rights reserved.

16

T.G. Nienhuys

CI al. / Int. J. Pediarr.

Olorhinolaryngol.

30 (1994)

15-27

tions. Such information is necessary when making recommendations about school students’ needs for hearing habilitation or special educational support. The first aim of this study, therefore, was to ascertain the range of hearing levels associated with different middle ear disease states. Numerous studies have attempted to define the severity and configuration of CHL associated with current middle ear disease in non-Aboriginal populations. For example, Fria et al. [12] found mean pure-tone averages (PTAs) and speech reception thresholds of 24.5 dB and 22.7 dB, respectively (range O-55 dB) in 540 children aged 2- 12 years with OM with effusion (OME). Kokko [ 141described the mean pure-tone audiogram of 146 patients (238 ears) with untreated chronic OME which showed an average hearing loss of 27.6 dB (S.D. 12.8) across test frequencies, with best acuity at 2 kHz (23.0 dB) and worst at 8 kHz (33.8 dB). Other studies have reported mean hearing impairment with OME in the range 25-45 dB [8,13,14]. Bess [l] accumulated findings from three studies representing a total of 627 ears [2,8,14] and reported that, while hearing levels associated with OME may range from normal to 50 dB, the majority of ears were in the range 21-30 dBHL. Hearing level has also been found to be related to the amount of fluid in the middle ear [2,26], while Roland et al, [20] have confirmed a significant relationship between hearing level and middle ear condition on the same day in their subjects. While previous studies have focused on non-Aboriginal children with OME (or secretory OM), no studies of non-Aboriginal children have described expected hearing levels with middle ear disease showing wet or dry perforated tympanic membranes (TMs) - chronic suppurative OM (CSOM) - as are frequently found among Australian Aborigines. Furthermore, previous studies of Aboriginal children have not related CHL levels to the middle ear disease state. For example, Stuart [24,25] has reported hearing levels (in one case under controlled acoustic conditions [24]), but not for OME and perforated TM conditions separately. A subsequent study [23] attempted a partial separation of middle ear states by grouping children (rather than ears) according to whether one or both TMs were perforated. This study reported mean PTAs (across 1, 2 and 4 kHz) of 21.9 and 26.5 dB in child groups showing OME and CSOM, respectively in at least one ear. While the actual range of hearing levels found was not reported, it appeared that ears with TM perforations showed worse hearing. A study of 179 Aboriginal subjects by Macintosh ,[171showed that ears with TM perforations showed significantly worse hearing than those with OME, although actual hearing levels for the various ear states were not reported. Finally, Clapin [6] reported a mean PTA hearing level of 35.9 dB for 80 ears with CSOM, although testing was not performed under controlled acoustic conditions and it is not clear whether ear discharge was removed before hearing was tested. The second aim of this study was to find which middle ear diagnostic method best predicted the presence of significant CHL. In remote Australian Aboriginal communities, audiometric training of clinic staff is usually very limited, equipment is difficult to maintain and calibrate, and, suitable acoustic conditions for audiometry are usually not available. Nevertheless, knowledge of schoolchildren’s current CHL is essential when deciding on special educational and audiologic needs and, in the context of Australian Aboriginal community schools with high prevalence of untreated

T.G. Nienhuys

et al. /Ini.

J. Pediatr.

Otorhinolaryngol.

30 (1994)

15-27

17

middle ear disease [ 181, a simple clinical procedure which reliably identifies children requiring formal audiologic assessment would be invaluable. In other populations, relationships between immittance testing, pure-tone audiometry and otological diagnoses of OME have been explored, although these have usually evaluated reliabilities of techniques for identification of middle ear fluid. For example, reduced TM mobility observed with Siegle’s pneumatic otoscope has been proposed as a reliable diagnostic sign of OME [14]. Bluestone et al. [2] compared immittance testing, otoscopy and pure-tone hearing testing as predictors of OME, the presence of which was verified during myringotomy. They found that only 79% of ears with confirmed effusion showed significant hearing loss (mean HL 2 25 dB). For immittance testing, effusions were found in all ears showing flat tympanograms but not at all in ears showing normal-to-high compliance. They concluded that tympanometry was more reliable and sensitive than hearing testing in identifying middle ear effusions, and ‘probably equal or superior in reliability to simple otoscopic examination’ (Ref. 2, p. 604). In a study of the reliability of screening procedures for middle ear disease, Renvall and Liden [ 191concluded that supplementing immittance testing with two-tone hearing screening would reduce the screening rate of false identifications of OME. Fewer studies have evaluated middle ear measures as predictors of significant CHL; Dempster and MacKenzie [9] showed that the presence of a peaked tympanogram was a highly sensitive predictor of normal hearing while children showing flat tympanograms should have their hearing tested directly to confirm significant CHL. No previous attempts have been made to compare reliabilities of middle ear diagnostic measures in the Australian Aboriginal population where middle ear disease is more prevalent, more severe (frequently including CSOM and other more severe middle ear pathologies), of longer duration and in remote area contexts of limited diagnostic skills, limited test equipment and little or no treatment. Therefore, this study aimed to compare the value of tympanometry and pneumotoscopy as predictors of significant CHL in a sample of Aboriginal schoolchildren under controlled acoustic conditions. Reflectometry was included and results describing its limited value with remote area Aboriginal populations are reported elsewhere [3]. 2. Methods 2.1. Subjects Subjects were all available children aged 4- 16 years attending an Aboriginal community school in the tropical north of the Northern Territory, Australia. The study took place in the month of August, near the end of the ‘dry season’. A total of 255 children (428 ears) aged up to 16 years were examined, 133 male and 122 female. The numbers of ears examined with each test procedure in four age groups varied according to availability of children for the test procedures as shown in Table 1. 2.2. Otoscopy

Independent otoscopic examinations were performed by a trainer (an experienced, pediatric otologist, J.V., or a medical officer, F.Mc.C., whose observations were previously validated against those of J.V.) and a trainee (an Aboriginal health

T.G. Nienhuys et al. /Int. J. Pediatr. Ororhinolaryngol. 30 (1994)

18

Table 1 Number of cars in four age groups examined by pneumotoscopy, audiometry

tympanometry

15-27

and pure-tone

Number of ears examined by:

Age (years)

otoscopy

Tympanometry

Audiometry

O-6 l-9 10-12 13+

110 121 III 86

83 103 96 74

16 110 92 82

Total

428

356

360

(E.G.), registered nurse (P.S.) or audiologist (J.B.)); only trainers’ observations are reported here. Canal contents were recorded as (1) clear or some wax, (2) discharge present, (3) wax removed, or (4) discharge removed. Static tympanic membrane appearance was recorded as (1) pristine, (2) uncertain, (3) brownish/fluid present, (4) perforated. TM mobility under pressure applied with a Siegle’s speculum was recorded as (1) mobile, (2) reduced mobility, (3) immobile, or (4) consistent with perforation. Myringotomy for confirmation of effusions did not gain Ethics Committee approval. worker

2.3. Immittance testing Middle ear immittance was tested with a GSI 27A Automatic Tympanometer on 356 ears without TM perforation. Normal peak pressure was taken as the range +lOO daPa to -99 daPa and normal compliance from 0.3 to 1.6 ml. Tympanograms were then classified as follows using an adaptation of categories suggested by Cantekin [4] and Cantekin et al. [5]: Type Type Type Type Type

Al A2 A3 B Cl

Type C2 Type C3

normal peak pressure; compliance within normal range normal peak pressure; high compliance (> 1.6 ml) normal peak pressure, low compliance (< 0.3 ml) no pressure/compliance peak pressure peak - 100 daPa to - 199 daPa, normal or high compliance (>0.2) pressure peak -200 daPa or less, normal or high compliance (> 0.2) pressure peak -100 daPa or less, low or rounded compliance (<0.3)

2.4. Pure-tone hearing testing Pure-tone air conduction audiometry was performed on 180 cooperative children (360 ears) by the audiologist or the nurse in a specially-constructed, mobile, soundattenuated test booth which satisfied Australian ambient noise standard AS 1269 [23]. A Grason Stadler GSI 16 diagnostic audiometer was used to test thresholds at

T.G. Nienhuxs et ul. / Int. J. Pediutr.

Otorhinolur~ngol.

30 (I9941

I’)

15-27

250 Hz, 500 Hz, I kHz, 2 kHz and 4 kHz. Pure-tone average hearing loss categories were defined according to Clark [7] based upon the test frequencies 500 Hz to 4 kHz. 3. Results 3.1. Otoscopy The majority of ear canals were clear; there was evidence of a discharging TM in lo”/;, of ears and pus was removed by irrigation and/or dry mopping from 3% of canals. The appearance of the TM was normal in about one-half of the ears and in-

r 0-ISyears

I

H

Normal mobility

a

Reduced mobility

H

Perforated

29.0%

37.3%

: :::::::: :::::: ::::.,:::::::::.:.: : : : : :‘:‘:‘:‘:

49.6%

12.4% 14.5% 38.0%

[0-]

j]

12.6%

13.5% [13years+]

[-TX&Z-T-

Fig. I. Otoscopic findings of tympanic

membrane

appearance

and mobility for 428 ears

20

T.G. Nirnhuys

et ul. /Int.

J. Pediatr.

Otorhinolaryngol.

30 (1994)

15-27

Table 2 Tympanometric types for ears in four age groups

Age

Tympanometric type

(years) Al

A2

A3

B

Cl

c2

c3

DNT

Total II0 121

O-6 7-9 IO-12

7 25 41

0 I 0

4 IO 9

54 42 22

I 6 I6

4 II 5

I 8 3

21 I8 IS

l3+

32

I

IO

I9

3

5

4

I2

10s 30

2
33 9

137 38

32 9

25 7

22 6

72 -

Total ‘%,

III 86

428

DNT, did not test.

tact but abnormal in 37%; 29% of intact TMs had reduced or minimal mobility when pressure was applied with the pneumatic otoscope. Proportions of TM appearance types and mobility on pneumotoscopy for all ears and for four age subgroups are shown in Fig. 1; percentages of normally mobile TMs increased with child age with corresponding reductions in proportions of TMs showing reduced mobility or perforation, supporting previous reports that Aboriginal schoolchildren’s middle ear condition improves with age into adolescence [17]. 3.2. Immittance testing Table 2 shows the frequency distribution of tympanometric types for ears with intact TMs and no other conditions which would contra-indicate tympanometry; data are shown for all ears tested and for four age subgroups. For all ears, flat (Type B) tympanograms were the most frequent (38% of ears), while only 30% of ears showed

Pure-tone average (dBHL)

Fig. 2. Distribution of pure tone average hearing levels for all ears, ears with Type Al tympanograms, ears with Type B tympanograms and ears with perforated tympanic membranes.

T.G. Nienhuys

et al. /Int.

J. Pediatr.

Otorhinolarwgol.

30 119941 15-27

?I

peak pressures and compliance within the normal range (Type Al). As with otoscopy, the table again shows improvement with child age as found in previous studies [ 171;for example, there are proportionally more Type A 1 tympanograms and fewer Type B tympanograms in children aged 10 years and over. 3.3. Pure-tone threshold hearing testing Fig. 2 shows the distribution of pure-tone average hearing thresholds for all 360 ears tested (average of 500 Hz, 1 kHz, 2 kHz and 4 kHz), for ears showing normal Type Al tympanograms, for ears showing Type B tympanograms and for ears with perforated tympanic membranes. The mean pure-tone average for ears with normal tympanograms was 11.2 dB (S.D. 5.9). The mean pure-tone average for all 360 tested ears was 17.6 dB (S.D. 10.0); as expected in a sample with high prevalence of middle ear disease, mean hearing acuity for all ears was best at 2 kHz (14.4 dB) and worst at 4 kHz and 500 Hz (20.3 dB and 17.6 dB respectively). Only 54% of ears showed hearing in the normal range (pure-tone average 5 15 dB, as defined by Clark [7]), and 16% showed pure-tone average hearing levels exceeding 25 dB; only one ear showed average hearing levels in the moderately severe range, greater than 55 dB. (Although pure-tone average hearing levels between 16 dB and 25 dB may be regarded as deleterious to speech and language development [2], 25 dB is considered here because it is the screening level commonly used by service providers with this population in Australia.) 3.4. Relationships between pneumotoscopy and immittance testing When results from pneumotoscopy and immittance testing in the same 356 ears were compared, it was found that in 78”/0of ears pneumotoscopy and tympanometry results agreed: in 51”/0of ears TMs were assessed as mobile on both pneumotoscopy and tympanometry and 27% of ears in which TM mobility was reduced or minimal (Categories 2 or 3) had Type B or C3 tympanograms. In the remaining 22% of ears, 18% had TMs categorised as mobile on pneumotoscopy but which showed reduced mobility on tympanometry (Type B or C3), and 4% showed reduced or minimal pneumotoscopic mobility but were mobile on tympanometry. It may be that when mobility was observed with otoscopy but not with tympanometry it was due to observations made under the greater pressure applied with Siegle’s speculum. 3.5. Relationships between pure-tone hearing thresholds and other measures The prevalence of pure-tone average hearing levels in excess of 25 dB was 25.3% among ears with Type B tympanograms; 30.0% for ears showing a Type B tympanogram and an immobile or reduced mobile TM under pneumotoscopy; 57.4% for ears with perforated TMs. Fig. 3 shows the mean audiograms (and standard deviations) for ears which showed Type B tympanograms and for those with perforated TMs. The data from all ears were analysed using the computer package Generalised Linear Interactive Modelling (GLIM) [22] to explore possible relationships between average pure-tone hearing levels, middle ear immittance, TM appearance and mobility, and subjects’ age and gender. Ears showing flat tympanograms had an average pure tone hearing level of 20.3 dB (S.D. 9.6), while those with perforations had an average hearing level of 30.0 dB (S.D. 11.1); mean pure-tone average hearing levels

T.G. Nienhuys et ul. /In!. J. Pediutr. Otorhinoluryngol. 30 (1994)

22

100' 500

1000

2000

IS-27

4000

Frequency (Hz)

Fig. 3. Mean audiograms

l

Mean hearing level (and standard deviations) Type B tympanogram

-

x

Mean hearing level (and standard deviations) perforated tympanic membranes

-

(and standard

deviations)

with perforated

for ears with Type B tympanograms

tympanic

and for those

membranes.

Table 3 Sensitivity, specificity and positive and negative predictive values (‘X) of simple otoscopy, pneumotoscopy and tympanometry as predictors of pure-tone average hearing level greater than 25 dB Diagnosis

1. 2.

TM appearance TM appearance

intact vs. perforated normal vs. any other category

4.

(including perforated TMs) TM appearance normal vs. any other category (excluding perforated TMs) TM mobility normal vs. reduced mobility or

5.

immobile (excluding perforated TM mobility normal vs. reduced

3.

TMs) mobility

or

Sens

Spec (‘%I)

PPV

NPV (‘%I)

47 89

93 60

51 29

90 97

80

65

25

97

80

74

25

97

89

69

35

91

46

62

I8

86

38 23 23

80 86 66

26 23 25

87 85 86

7. 8. 9.

immobile (including perforated TMs) TM appearance or mobility normal vs. abnormal TM appearance and mobility normal vs. abnormal TM mobility or appearance uncertain vs. abnormal TM mobility and appearance normal or uncertain

IO. I I.

vs. both abnormal Peaked typanogram vs. flat (Type B) tympanogram TM reduced mobility or immobile or Type B

90 49

68 62

26 20

98 87

12.

tympanogram vs. all others TM reduced mobility or immobile

40

83

32

88

6.

tympanogram PPV, positive

predictive

and Type B

vs. all others value; NPV, negative

predictive

value.

T.G. Nienhuys et al. /Int. J. Pediatr. Otorhinoiuryngol. 30 (19941 15-27

23

both for ears with perforated TMs and for those with intact, immobile TMs differed significantly from normal (P < 0.001). For mean pure-tone average hearing levels, no significant differences were found between males and females in the sample, while average hearing levels did signiticantly improve with subject age (P < 0.01). The number of TMs showing reduced mobility, immobility or perforation significantly reduced with increasing child age (P < 0.01) as did those showing Type B tympanograms (P < 0.01). When these effects of tympanic mobility were partialled out, mean hearing levels were no longer

20 ears

180

ears

6 ears

96

ears

24 e3s

20

205

ears

6 ears

73 ears

27 ears

27

24 ears

Fig. 4. Decision flow charts for referral for hearing assessment: protocol A, eardrum appearance only: protocol B. eardrum appearance and otoscopic mobility; protocol C. eardrum appearance and tympanometric mobility.

T.G. Nienhuys et al. /Inr. J. Pediarr. Otarhinolaryngol. 30 (1994)

188

ears

27

90

23

15-27

ears ears ears TP -TruePositive; TN -TrueNegative; FP - False Positive; FN - False Negative; DNT - did not test; TMs - tympanic membranes; Appear. - eardrum appearance with static otoscopy; PT - pure-tone average.

Diagnostic

TP + TN

FP

FN

A

64%

33%

2%

B

71%

26%

2%

C

68%

31%

1%

Protocol

Fig. 4 (continued).

significantly related to child age, suggesting that the age-related improvement in middle ear condition (as reflected in TM mobility, tympanometry and perforation) accounted for the better hearing in older groups. 3.6.‘ Sensitivity and specificity and predictive values For all ears receiving pure-tone audiometry, the efficiencies of otoscopic appearance, pneumotoscopy and tympanometry in screening for average hearing loss > 25 dB were determined, as shown in Table 3. When all ears, including those with perforated TMs were considered, sensitivity was highest for abnormal otoscopic appearance (89%) and for abnormal TM mobility (89%). When ears with perforated TMs were excluded, Type B tympanograms detected significant hearing loss with highest sensitivity (90%). Positive predictive value was best for the diagnostic distinction of intact versus perforated TMs, because of the very high prevalence of hearing loss amongst ears with perforated TMs (57.4%). Fig. 4 shows three possible decision flow charts for referral for formal hearing test-

T.G. Nienhuys et al. /Int. J. Pediatr. Otorhinolar~ngol. 30 i 19941 IS-27

‘5

ing for three diagnostic protocols applied to the data from this study: protocol A, using static otoscopic appearance only; protocol B, using otoscopic appearance and TM mobility under pneumotoscopy; and protocol C, using simple otoscopic appearance and tympanometry. The figure shows that diagnostic protocol B optimised correct identification of significant hearing impairment and normal hearing with a minimal false positive rate and a comparable false negative rate in this population. 4. Discussion This study has a number of significant implications for the detection and educational management of Aboriginal students with conductive hearing loss associated with middle ear disease. This is the first reported study of OM-related conductive hearing loss levels measured under controlled acoustic conditions in a large remote area sample of Aboriginal schoolstudents. For ears showing OME (Type B tympanograms), the average of 20.3 dB and range of pure-tone hearing levels accorded well with earlier reports from non-Aboriginal samples [ 1,8,12,13,15,27] and with one study of Aboriginal students [24]. For ears with perforated TMs, the mean pure-tone average hearing level of 30.0 dB differed from previous reports: Lewis et al. [16] reported a mean hearing level of 26.5 dB which may be explained by their exclusion of 500 Hz test frequency, while Clapin [6] reported 35.9 dB which may be because she did not report removal of purulent discharge before threshold testing, and neither study tested hearing under controlled acoustic conditions. Despite the range of hearing levels associated with different middle ear states, the observed difference in mean hearing levels between the two groups may have significant educational and treatment implications. Firstly, children with perforated TMs are likely to suffer more severe hearing disability and to have a greater need for personal amplification in the classroom, whereas those with OME may still require personal amplification but may derive benefit from classroom strategies such as classroom noise control, daily nose-blowing and Valsalva manoeuvre, improved teacher communication techniques or low-level classroom amplification systems such as the MARRS system [28]. Even so, accurate fitting of personal amplification for children with TM perforation still awaits a better understanding of possible short-term (even daily) hearing fluctuations associated, say, with the presence of purulent discharge and size of TM perforation. Clapin [6], for example, found that Aboriginal children with persistent perforation had a greater average hearing fluctuation (15.8 dB) than did children with consistently normal middle ear function (10.9 dB) or intermittent middle ear dysfunction (10.1 dB) when tested monthly over a 1 year period. Secondly, these results suggest that medical treatment strategies for CSOM may usefully improve hearing levels even if the treatment results only in OME without totally restoring normal middle ear state and hearing function [14]. That is, even if treatment of CSOM results only in improving hearing levels, on average, by 10 dB because of residual OME, this improvement may be important to the child’s language acquisition and classroom learning, especially for Australian Aboriginal students, many of whom have not yet learned standard English which is used for

26

T.G. Nienhuys

et al. /Inr.

J. P&air.

Otorhinolaryngol.

30 (1994)

15-27

classroom instruction. Ideally, early and effective medical treatment would return all middle ears to a healthy state. However, this remote community has a high prevalence of respiratory disease and limited access to medical and surgical services. Therefore, the initial aim is to reduce the prevalence of ear discharge and encourage closure of perforations. The finding of improved hearing levels and middle ear condition with child age agrees with previous reports of higher prevalence and greater severity of middle ear disease (especially TM perforations) among Aboriginal infants and pre-school children and of some spontaneous middle ear improvement in older schoolchildren [ 10,171. Nevertheless, this improvement may well come too late for the younger children who are establishing early English oral and literacy skills, and medical strategies which improve their hearing are likely to be of greatest benefit to them. The study also found that pneumotoscopy for detection of middle ear effusion and perforated TMs resulted in the best rate of prediction of significant conductive hearing loss (requiring subsequent pure-tone audiometry) and a false negative rate of only 2%. This suggests that, for this population in remote locations with limited availability of trained personnel and calibrated audiometric or tympanometric equipment, pneumotoscopy is a simple and satisfactory method for detecting students who require full audiometric follow up. (Of course, all children should be tested with pure-tone audiometry at least once early in their lives to eliminate the possibility of sensorineural hearing disorders.) At present, however, doctors, nurses and Aboriginal Health Workers in remote clinics usually have only rudimentary skills in simple otoscopy, and TM mobility is not usually checked with Siegle’s speculum at all. In conclusion, this study suggests that, if training improved the pneumotoscopy skills of remote community clinic personnel to the levels of the trainers in this study, then pneumotoscopy would be valuable as a simple and effcient screening method for Aboriginal infants.and school students with significant conductive hearing loss. 5. Acknowledgements The authors wish to thank the children, their schoolteachers and staff of their community’s clinic who participated in or assisted with this study. We are also grateful to the otologist, J. Vorrath, and to E. Gilder and P. Simms for assistance with field trips and Aboriginal community liaison, and to Professor J. Mathews. This study was supported by grants from the National Health and Medical Research Council of Australia and the Australian Mutual Provident Society Medical Research Fund. 6. References 1 2 3

Bess, F.H. (1985) The minimally hearing-impaired child. Ear Hear. 6. 43-47. Bluestone, C.D, Beery, Q.C. and Paradise, J.L. (1973) Audiometry and tympanometry in relation to middle ear effusions in children. Laryngoscope 83. 594-604. Boswell, J.B. and Nienhuys, T.G. (1993) Reflectometric screening for otitis media: inconsistencies in a sample of Australian Aboriginal children. Int. J. Pediatr. Otorhinolaryngol. 25, 49-60.

T.G. Nirnhuys

4

5 6 7 8 9 10

II

I? 13

I4 I5 I6 I7 I8 19 20 21

22 23 24 25 26 27 28

Cantekin. Laryngol. Cantekin.

ef al. / Int. J. Pediotr.

Otorhinoluryn,qoL

30

i 1994 I

IS-27

17

E.1. (1983) Algorithm for diagnosis of otitis media with effusion. Ann. Otol. Rhinol. 92, (Suppl. 197) 6. E.I.. Bluestone. C.D.. Fria. T.J.. Stool SE., Beery. Q.C. and Sabo. D.L. (lY80) Identifica-

tion of otitis media with effusion in children. Ann. Otol. Rhinol. Laryngol. 89 (Suppl. 68). 190-195. Clapin, M. (1992) Hearing Thresholds in Western Australian Primary School Children. MAud Thesis. University of Queensland. Clark. J.G. (1981) Uses and abuses of hearing loss classification. A.S.H.A. 23. 493-500. Cohen, D. and Sade, J. (1972) Hearing in secretory otitis media. Can. J. Otolaryngol. I. ?7-2Y. Dempster. J.H. and Mackenzie, K. (1991) Tympanometry in the detection of hearing impairments associated with otitis media with effusion. Clin. Otolaryngol. 16. 157-l 59. Dugdale A.E., Canty. A.. Lewis. A.N. and Lovell. S. (1978) The natural history of chrome middle ear disease in Australian Aboriginals: a cross-sectional study. Med. J. Aust. Spec. Suppl. I. 6-Y. Fiellau-Nokolajsen, M. (1984) Tympanometric prediction of hearing loss in secretory otitis media. In: Lim. D.J.. Bluestone. C.D., Klein, J.O. and Nelson, J.D. (Eds.), Recent Advances in Otitis Media with Effusion, BC Decker Inc.. Philadelphia, pp. 336-329. Fria, T.J.. Cantekin. E.I. and Eichler, J.A. ( 1985) Hearing acuity of children with otitis media with effusion. Arch. Otolaryngol. I I I. 10-16. Fria. T.J.. Cantekin. E.L. Eichler, J.A.. Mandel. E.M. and Bluestone. C.D. (1984) In: Lim, D.J.. Bluestone. C.D.. Klein. J.O. and Nelson. J.D. (Eds.). Recent Advances in Otitia Media with Effusion, BC Decker Inc., Philadelphia. Giebink. G.S. and Daly. K. (1990) Epidemiology and management of otitis media in children. Top. Lang. Disord. II (I), I-IO. Kokko. E. (1974) Chronic secretory otitis media in children. Acta Otolarynogol. Suppl. 327. l-44. Lewis. A.N., Barry, M. and Stuart, J.E. (1974) Screening procedures for the identification of hearing and ear disorders in Australian Aboriginal children. J. Laryngol. Otol. 88 (4), 335-347. McCafferty, G.J.. Lewis, A.N., Coman. W.B. and Mills. C.A. (1985) A nine-year study of ear disease in Australian Aboriginal children. J. Laryngol. Otol. 99. 117-125. McIntosh. E.D.G. (1984) Factors preventing the spontaneous healing of perforated tympanic membranes in Aborigines. Master of Public Health treatise. University of Sydney. McPherson, B. (1990) Hearing loss in Australian Aborigines, Aust. J. Audiol. 12. 67-78. Renvall. U. and Liden, G. (1980) Screening procedure for detection of middle ear and cochlear disease. Ann. Otol. Rhinol. Laryngol. 4, 214-216. Roland. P.. Finitzo. T., Friel-Patti, S.D.. Clinton-Brown, K.C.. Stephens, K.T.. Brown, 0. and Coleman J.M. (1989) Otitis media: incidence. duration and hearing status. Arch, Otolaryngol. 115. 1049-1053. Royal Statistical Society: GLIM. (1987) Generalised interactive modelling. Release 3.77. Edition 2 Numerical Algorithms Group. Standards Association of Australia: AS1269 (1989) Hearing Conservation. Standards House. Sydney. Stuart. J.E. (1975) Follow-up studies of Aboriginal children with ear disease and hearing loss at Cherbourg (Queensland). Med. J. Aust. Spec. Suppl. I. 3X-40. Stuart, J.E., Quayle. C.J. and Lewis, A.N. (1971) Hearing and ear disease in Aboriginal schoolchildren. Proceedings of the International Congress on Pediatrics. Vienna, Div. 28. 147- 151. Stuart, J.E., Quayle. C.J., Lewis, A.N. and Harper. J. (1972) Health. hearing and ear disease in Aboriginal schoolchildren. Med. J. Aust. I, 855-859. Wiederhold, M.L., Zatjtchuk. J.T.. Vap. J.G. and Paggi, R.E. (1980) Hearing loss in relation to physical properties of middle ear effusions. Ann. Otol. Rhinol. Laryngol. Suppl. 69. 185-189. Worner. W.A. (1988) An inexpensive 90 ( 1). 29-36.

group

FM amplificatton

system for the classroom.

Volta Rev.