Quality Control of CE-Certified Phonak Hearing Aids - 2020_43

A listening effort scaling procedure ('ACALES'; Krueger et al, 2017) was performed with noise cancellation on and off. ACALES is adaptive, altering the signal-to-noise ratio (SNR) at which speech is presented. Therefore it is not a simple rating scale. For each stimulus, a rating (14-point scale ranging from 1='effortless' to 14= 'only noise') was given for how effortful it was following speech. This results in 2-D data: a list of presentation SNRs (different for each participant), and respective participant ratings. Given the format of this data, it cannot be summarized using standard descriptive statistics. We defined SNR as the dependent variable and fitted a linear mixed effects model, with the predictors: listening effort rating & intervention (treating participants as a random factor), characterizing SNR over the entire rating scale. The values reported represent the estimated marginal means of SNR for each intervention, averaged across effort ratings & participants.

Time frame: 2 weeks

Population: In total there were 8 participants, who performed this test. It was a crossover- design, so each participant completed both arms during a single appointment in a randomized order.~9 participants started the test, but 1 was not able to perform the test, so the procedure was stopped for this participant and no data was collected.

This outcome measure evaluated the ability to detect sounds in a quiet environment. A comparison was made between the feature under investigation being activated and deactivated. Participants were asked to change the level of a speech sample until they were just able to detect it. The ability will be measured by a detection threshold (in dB).

Time frame: 2 weeks

The secondary Outcome measure of this study contains 2 parts. This first part reported here was the evaluation of the sound quality. A comparison was made between speech and music audio excerpts either with or without gain being increased by the feedback manager. Subjective ratings of sound quality were made under both of these conditions for one speech and one music sample, using the following scale: Bad (1) - Poor (2) - Fair (3) - Good (4) - Excellent (5). Results were analyzed separately for the speech and music samples.

Time frame: 4 weeks

Population: The test was done by 9 participants, but just the results of 8 participants could be used for the analysis.~This is because technical limitations meant it was not possible to perform the test under both experimental conditions for one participant.

The second part of this secondary Outcome measure was the evaluation of audible artefacts either with or without the available gain being increased by the feedback manager. The outcome was measured as a binary response (artefacts audible/ not audible). Participants had to indicate whether artefacts were audible under both of these conditions. The number of positive ratings (artefacts were audible) was counted.

Time frame: 2 week

Population: 9 participants completed this study period, but just the results of 8 participants could be used.~This was because technical issues limited the reliability of the measures made for one participant.~N.B. this test was independent from the other tests in the study. Therefore the results are not influenced by other test periods. The number of people included in the analysis described here relates only to this test.