Transcript Children with Mild and Unilateral Hearing Impairment

EDHI Feb 2004
Children with Mild and
Unilateral Hearing
Impairment
Current management and
outcome measures
Kirsti Reeve Ph.D.
Developmental Disabilities Institute
Wayne State University, Detroit MI
Overview
Current management for children with mild
and unilateral hearing impairment (HI)
 Outcome measures:

 Speech
and language
 Cognition
Why these populations?
Very little known about management or
outcomes for mild or unilateral HI
 NHSP offers the potential for early
identification
 There is strong evidence that early
identification gives improved outcomes in
moderate and greater HI populations
 Need to ascertain whether it would be
appropriate for these groups

Study overview

Two separate studies
 Questionnaire
survey to audiologists
investigating management options
 Outcomes study
Obtained epidemiological data
 Assessed impact of HI quality of life
 Assessed impact of HI on speech, language &
cognition

Current management:
options for children with mild or
unilateral hearing impairment
Why assess service provision?

Areas of uncertainty
 Numbers
of children being identified
 Age of identification
 Management options for these groups
 Level at which to provide hearing aids
How was it done?
Single page questionnaire survey
 Sent out to 131 professionals
throughout the UK
 1 reminder
 56 responses (43%)

Results

Information on the mild and unilateral
cases seen

Management offered to those cases
Results

Information on the mild and unilateral
cases seen

Management offered to those cases
Numbers of children with
bilateral mild impairment

Defined as 20-40dBHL permanent
sensorineural loss

Comprise 8% of total caseload

Range seen from 0 to 300 (mean of 25)

Estimated total number seen by 56
clinicians: 1220
Numbers of children with
unilateral hearing impairment

Defined as permanent sensorineural
loss in one ear only.

Comprise 4% of total caseload

Range seen from 0 to 40 (mean of 9)

Estimated total number seen by 56
clinicians: 443
Numbers of children seen
N
25
Mild
Unilateral
20
15
10
5
0
0
1-5
6-10
11-15
16-20
Numbers
21-25
26-30
31-35
36+
Numbers found

Literature estimates prevalence figures at:
 between 0.5-5.2% for unilateral impairment
 Between 1 and 5.4% for mild impairment

“It is well recognized that an inverse relationship
exists between the prevalence and degree of
hearing loss” – Bess 1984

The low percentage as ascertained by this
questionnaire would imply that large numbers of
these populations are not receiving audiological
management
Age of Referral
N
35
Mild
30
Unilateral
25
20
15
10
5
0
0-6 months 6-12 months
1-3 years
Age
4-6 years
6 years +
Age of referral, data from Trent Region
(Mild n=50, Unilateral = 30)
percentiles
100
90


80

70


50

40

30
 

X

X







60
X
X
X
Moderate
 All Trent
 Mild
X Unilateral

X
,
20

,
10

,
0 ,)'
0
10
20
30
40
50
60
70
80
Age in months
90
100
110
120
130
140
Age of referral
Age of referral is late for both groups of
children when compared with Trent data
 Children with unilateral impairment
identified significantly later than children
with mild impairment
 Modal age of between 4 and 6 years
suggests that the school entry is a factor
leading to identification

Results

Information on the mild and unilateral
cases seen

Management offered to those cases
Management
Unilateral
None/Discharge
Mild
Options
Speech Therapy
Review
Refer
Advice
Aid
0
10
20
30
Frequency
40
50
Management

Most frequent options are review and
advice

Children with mild HI are significantly more
likely to be offered:
 Hearing Aids
(p=0.0005)
 Speech Therapy (p=0.003)
 Referral to other professional (p=0.022)
Provision of aids
Uncertainty among professionals on
whether to aid mild HI
 Level below which you would not consider
providing aids:

 25dBHL

(range from 15 - 35dBHL)
Level above which you would definitely
provide aids:
 40dBHL
(range from 25 to 50dbHL)
Management conclusions
Mild and unilateral HI are underrepresented in the caseload of this sample
 These groups of children are identified
later than children with more severe
impairments
 Management is still uncertain whether to
provide aids and at what level for children
with a bilateral mild impairment

Outcome Measures –
Language and Cognition
Hypotheses

Language is likely to be affected to some
degree by a mild or unilateral hearing
impairment

There will be a positive relationship
between language scores, non-word
repetition and verbal reasoning
Subjects

41 children from CHAC met study criteria:
 Aged
6-11
 Bilateral mild, or unilateral hearing impairment
 HI is sensorineural
 No associated syndromes, or other problems.
 No known learning or cognitive disabilities.
 English as first language

20 children agreed to participate though one
child DNA’d twice, and was not followed up a
third time.
Participants


8 mild
11 unilateral
6
mild, 3 moderate, 1 severe, 1 profound
 5 left ear impaired, 6 right ear



5 girls, 14 boys
Aged 6-11, average age 8yrs 3 months
Age of identification ranged from 9 months to 6
years 7 months (mean of 2 years 4 months)
Assessments

The session consisted of:
 Computer
based test of sound lateralization
 Standardised language assessment (CELF-3
UK)
 Children’s test of Non word Repetition
 BAS verbal & non-verbal reasoning (IQ)

Most sessions lasted 90-120 mins
including breaks.
Results

Language

Non-word repetition

Cognition
Results

Language

Non-word repetition

Cognition
Language testing - CELF 3 UK

Standardised on UK population
 Six

subtests:
3 for receptive language (understanding)
Sentence Structure (aged 6-8) / Semantic
Relationships (aged 9+)
 Concepts and Directions
 Word Classes


3 for expressive language (speaking)
Word Structure (6-8) / Sentence Assembly (9+)
 Formulated Sentences
 Recalling Sentences

Means of all language scores
Standardised Test: mean:100, sd:15
110
Receptive Language
Mean: 89.65,
sd 13.18
100
Expressive Language
Mean: 85.76,
sd 13.51
90
Total Language
Mean: 86.29,
sd 14.01
95% CI
80
70
N=
17
17
Receptive language s
17
Total language score
Expressive language
Total Language Scores for
individual subjects
120
110
100
90
80
70
60
.5
Unilateral HI
1.0
ty pe o f hearing i mpairment
1.5
Mild HI
2.0
2.5
Speech & language results 1

Unilateral group - total language score
mean of 91.78

Mild group - total language score mean of
80.12

With a linear regression, the difference in
scores just misses significance (.089) this could be due to the small sample size.
Speech & language results 2

Converting scores to age equivalent gives
an average language delays of:
6
months for children with a unilateral
impairment
 24 months for children with a mild impairment
Results

Language

Non-word repetition

Cognition
Children’s Test of Non-word
Repetition
Assesses phonological memory, and is
predicative of literacy development
 Administered via computer


Scores converted to standard scores,
with a mean of 100, sd of 10
CN-Rep Results 1

Both groups of children scored below
100 on this task
 Children
with mild HI: mean= 87.75
 Children
with unilateral HI: mean=95.55
CN-Rep results by type of hearing
impairment
110
100
90
80
70
N=
11
8
unilateral
bilateral
type of hearing impairment
CN-Rep results 2

Significant correlation of .953 with the
recalling sentence CELF subtest (p=0.005)
controlling for age

Scores can be compared with those from
an OME group and hearing controls from
BOS study
CN-Rep scores as a function of type of
hearing impairment
115
110.7
110
105.2
standard score
105
100
95.6
95
87.8
90
85
80
75
70
None
OME
Unilateral
type of hearing impairment
Mild
Results

Language

Non-word repetition

Cognition
Cognition
Two tests from the British Abilities Scale
(BAS)
 Similarities (verbal reasoning)

 Why

do these things go together:
“milk, lemonade, coffee” , “cod, shark, pilchard”
 Need

to produce the superordinate
Matrices (non verbal reasoning)
 Finish
the pattern
Cognition results 1

Similarities (verbal reasoning)
 centile
scores ranged from 17-84
 mean of 45.71, sd 20

Matrices (non verbal reasoning)
 centile
scores ranged from 29-99
 mean of 77.82, sd of 23.55

So - significantly impaired scores on verbal
reasoning (p<.001 on independent samples ttest)
Cognition results 2

Only 3 children, all with mild HI, had higher
verbal than non-verbal reasoning

Mean difference of 32 centiles between
verbal and non-verbal scores

Significant difference in non-verbal score
depending on type of HI
 Independent
samples t test gives p=0.027
Cognition results
100
80
60
40
95% CI
Similarities centile
matrices centile
20
N=
9
9
unilateral
type of hearing impairment
8
8
bilateral
Cognition results 4

Correlation of .625 between verbal
reasoning and CELF language scores
(p=0.003)

Results can be compared across severity
range with outcomes data from larger
studies
Reasoning scores as a function of type
of hearing impairment
70
Verbal
60
t-score
50
40
30
20
10
0
ou
nd
re
e
at
er
ve
of
Pr
Se
od
M
ild
M
al
er
t
ila
Un
e
E
M
O
n
No
Reasoning scores as a function of type
of hearing impairment
70
Verbal
60
t-score
50
40
30
20
10
0
ou
nd
re
e
at
er
ve
of
Pr
Se
od
M
ild
M
al
er
t
ila
Un
e
E
M
O
n
No
Outcome measures conclusions
The caveat  These children were all identified through
CHAC. Therefore they have made it to the
attention of the audiology services
 There may be ascertainment bias which
could effect the results and make
generalisation more difficult
Outcome measures conclusions

Laterality of impairment for the unilateral
group was not predictive of performance

Greater severity of impairment was
correlated with better performance on
language outcomes ...

… although numbers are very small
Outcome measures conclusions

Children with mild or unilateral hearing HI who
are known to audiology services could be at risk
for developing language problems

Children with a bilateral mild impairment are
perhaps at greater risk than those with a
unilateral impairment, regardless of severity
Overall conclusions
Children with mild impairments are
showing language deficits that may
possibly be ameliorated through earlier
identification
 There is a need for further research in the
area of amplification provision for mild
impairments

Overall conclusions
Children with unilateral impairments
showed a variable performance which was
not correlated with severity of impairment
or side of impairment
 There is the need for a larger study to
investigate these findings further
 Early identification through NHSP is still
recommended

Thanks to
Adrian Davis and Sally Hind at MRC
Institute of Hearing Research, Nottingham
 Paul Shaw and the staff at CHAC,
Nottingham
 Helen Spencer & Jabulani Sithole for
statistical assistance
 Medical Research Council for PhD funding

Reasoning scores as a function of type
of hearing impairment
70
Non Verbal
60
t-score
50
40
30
20
10
0
ou
nd
re
e
at
er
ve
of
Pr
Se
od
M
ild
M
al
er
t
ila
Un
e
E
M
O
n
No
Reasoning scores as a function of type
of hearing impairment
64
Non Verbal
62
60
t-score
58
56
54
52
50
48
46
ou
nd
re
e
at
er
ve
of
Pr
Se
od
M
ild
M
al
er
t
ila
Un
e
E
M
O
n
No
Reasoning scores as a function of type
of hearing impairment
Verbal
Non Verbal
70
60
t-score
50
40
30
20
10
0
ou
nd
e
l
ra
t
ra
re
ve
of
Pr
Se
e
od
te
la
E
e
ild
M
M
ni
U
M
O
on
N
Reasoning scores as a function of type
of hearing impairment
70
Verbal
Non Verbal
60
t-score
50
40
30
20
10
0
ou
nd
e
l
ra
t
ra
re
ve
of
Pr
Se
e
od
te
la
E
e
ild
M
M
ni
U
M
O
on
N
Why does the prevalence
increase with age?

Is it…
 new
cases (i.e. acquired losses)?
 progressive
 late
nature of mild cases?
onset?
 persistent
OME?