Comparison of Different Hearing Aid Prescriptions for Children

Objectives: To assess whether there are significant differences between
speech scores for different hearing aid prescription methods, specifically
DSL i/o, DSL V, and NAL-NL1, using age-appropriate closed- and
open-set speech tests with young children, designed to avoid floor and
ceiling effects.
Design: Participants were 44 children with moderate or severe bilateral
hearing loss, 8 aged 2 to 3 years, 15 aged 4 to 5 years, and 21 aged 6
to 9 years. Children wore bilateral hearing aids fitted with each prescription
method in turn in a balanced double-blind design. The speech tests
used with each child (and for some tests the levels) were chosen so as
to avoid floor and ceiling effects. For the closed-set tests, the level used
was selected for each child based on their hearing loss. The tests used
were: (1) The closed-set Consonant Confusion Test of word identification;
(2) The closed-set Chear Auditory Perception Test (CAPT) of word
identification. This has separate sections assessing discrimination of
consonants and vowels and detection of consonants; (3) The open-set
Cambridge Auditory Word Lists for testing word identification at levels of
50 and 65 dBA, utilizing 10 consonant–vowel–consonant real words that
are likely to be familiar to children aged 3 years or older; (4) The openset
Common Phrases Test to measure the speech reception threshold
in quiet; (5) Measurement of the levels required for identification of the
Ling 5 sounds, using a recording of the sounds made at the University
of Western Ontario.
Results: Scores for the Consonant Confusion Test and CAPT consonant
discrimination and consonant detection were lower for the NAL-NL1
prescription than for the DSL prescriptions. Scores for the CAPT vowelin-
noise discrimination test were higher for DSL V than for either of the
other prescriptions. Scores for the Cambridge Auditory Word Lists did
not differ across prescriptions for the level of 65 dBA, but were lower
for the NAL-NL1 prescription than for either of the DSL prescriptions for
the level of 50 dBA. The speech reception threshold measured using the
Common Phrases Test and the levels required for identification of the
Ling 5 sounds were higher (worse) for the NAL-NL1 prescription than
for the DSL prescriptions.
Conclusions: The higher gains prescribed by the DSL i/o and DSL V prescription
methods relative to NAL-NL1 led to significantly better detection
and discrimination of low-level speech sounds.