Outcome measures
The Health Utilities Index Mark 3 (HUI3), a generic measure of health
utility gains (Feeny et al., 1995) was the primary measure. Change from
baseline to first follow-up in the HUI multi-attribute above or below
0.1 units was the primary outcome in this study.
Table 1 illustrates the structure of the HUI3 hearing attribute levels
(Feeny et al., 2012). Levels are designated based on the combined
responses to two questions about hearing. Levels 1 and 2, as the
highest/best levels indicate that the respondent does not always need a
hearing aid to hear what is said, whereas the lower levels of
performance 3-5 require use of a hearing aid (or other assistive hearing
device) and report the level of performance with that device. Level 6
indicates unable to hear anything at all, even with hearing aids.
Furthermore, only levels 1-3 indicate that respondents can hear what is
said in more difficult situations. We have designated the categories
“able to hear” and “unable to hear” to the hearing attribute levels
1-3 and 4-6 for the purposes of the discussion. Obtaining level 3 (or
greater) means that the intervention was essentially successful in
restoring the capacity to hear what is said, even in more challenging
situations, so long as a hearing aid, or in this case CI, is being used.
Weighting scores for given levels of each health attribute are used to
produce a HUI3 “total” or multi-attribute score combining all eight.
As in table 1, some transitions between adjacent hearing attribute
levels (e.g. for hearing 6 to 5, and 4 to 3) may possibly result in
changes of ≥0.1 in the combined HUI3 multi-attribute scores, but often
at least two levels need to be gained for this to be the case. This is a
property of the way the HUI3 was constructed.
Data were also collected for the Speech Spatial Qualities (SSQ) scale
(Gatehouse and Noble, 2004). This is a disease specific scale which aims
to evaluate the subject’s speech understanding in quiet and noise,
spatial perception and the clarity, separation, and identification of
sounds. Changes of at least 1.0 units on the SSQ subscales indicates a
clinically relevant change (Noble and Gatehouse, 2006). Additional
information was collected about each subject’s hearing history and
demographics.
TABLE 1 HERE